Skip to content

alexgreensh/repo-forensics / skills/repo-forensics

repo-forensics

Flagged

Security forensics for git repos, AI skills, and MCP servers. Audits dependencies, detects prompt injection, credential theft, runtime dynamism, manifest drift, known CVEs, CISA KEV (actively exploited) vulns, and 2026 attack patterns. Not for fixing vulnerabilities or pentesting.

FileKindSize
.gitignoreother193 B
SKILL.mdskill md43.4 KB
checksums.jsonconfig10.8 KB
checksums.json.sigother64 B
data/README.mdmarkdown1.3 KB
data/compromised_versions.jsonconfig188.3 KB
data/rule_ids.csvother35.3 KB
data/rulepacks/.gitkeepother221 B
data/rulepacks/dead_anchors.jsonconfig9.0 KB
data/rulepacks/mcp_security.jsonconfig31.2 KB
data/rulepacks/runtime_dynamism.jsonconfig46.7 KB
data/rulepacks/sast.jsonconfig107.6 KB
data/rulepacks/secrets.jsonconfig30.4 KB
data/rulepacks/shared.jsonconfig4.4 KB
data/rulepacks/skill_threats.jsonconfig95.4 KB
data/runtime_floors.jsonconfig2.3 KB
data/yara/cryptominers.yarother1.8 KB
data/yara/hacktools.yarother2.2 KB
data/yara/malware.yarother2.2 KB
data/yara/manifest.jsonconfig8.0 KB
data/yara/webshells.yarother2.5 KB
references/cursor-hook-contract.mdmarkdown5.6 KB
references/kernel_privesc_sandbox_pack.mdmarkdown5.0 KB
references/mcp-attack-patterns.mdmarkdown8.7 KB
references/research_sources.mdmarkdown8.0 KB
references/threat_patterns.mdmarkdown20.3 KB
scripts/_context_gate.pyscript21.7 KB
scripts/_ed25519.pyscript5.1 KB
scripts/_pyc_unmarshal.pyscript17.1 KB
scripts/_shared_patterns.pyscript4.8 KB
scripts/adjudication.pyscript16.4 KB
scripts/adjudication_bridge.pyscript1.0 KB
scripts/aggregate_json.pyscript69.2 KB
scripts/auto_scan.pyexecutablescript152.1 KB
scripts/corpus_sync.pyscript10.0 KB
scripts/cursor_pre_scan.pyscript22.5 KB
scripts/dast_sandbox.sbother1.6 KB
scripts/dead_anchors_extract.pyscript14.8 KB
scripts/dead_anchors_probe.pyscript27.1 KB
scripts/forensics_core.pyscript176.5 KB
scripts/gen_rule_ids.pyscript8.6 KB
scripts/hook_adapter.pyscript19.4 KB
scripts/ioc_manager.pyscript49.6 KB
scripts/parse_pnpm_lock.pyscript7.9 KB
scripts/pre_scan.pyscript143.5 KB
scripts/pypi_pins.pyscript8.1 KB
scripts/refresh_controller.pyexecutablescript48.4 KB
scripts/refresh_threat_dbs.pyexecutablescript19.4 KB
scripts/rule_loader.pyscript44.8 KB
scripts/rulepack_feed.pyscript26.4 KB
scripts/run_forensics.shexecutablescript20.3 KB
scripts/sarif_export.pyscript18.6 KB
scripts/scan_agent_skills.pyscript33.2 KB
scripts/scan_archive.pyscript29.1 KB
scripts/scan_ast.pyscript44.9 KB
scripts/scan_binary.pyscript10.9 KB
scripts/scan_bytecode.pyscript28.2 KB
scripts/scan_dast.pyscript28.8 KB
scripts/scan_dataflow.pyscript15.4 KB
scripts/scan_dead_anchors.pyscript15.9 KB
scripts/scan_decode.pyscript37.5 KB
scripts/scan_dependencies.pyscript124.5 KB
scripts/scan_devcontainer.pyscript8.5 KB
scripts/scan_entropy.pyscript15.6 KB
scripts/scan_entrypoint.pyscript34.3 KB
scripts/scan_git_config.pyscript96.3 KB
scripts/scan_git_forensics.pyscript24.1 KB
scripts/scan_history.pyscript14.5 KB
scripts/scan_infra.pyscript54.3 KB
scripts/scan_integrity.pyscript41.0 KB
scripts/scan_lifecycle.pyscript65.1 KB
scripts/scan_manifest_drift.pyscript47.0 KB
scripts/scan_mcp_security.pyscript42.3 KB
scripts/scan_oversize.pyscript11.7 KB
scripts/scan_post_incident.pyscript36.9 KB
scripts/scan_provenance.pyscript19.4 KB
scripts/scan_runtime_dynamism.pyscript23.5 KB
scripts/scan_runtime_floors.pyscript33.5 KB
scripts/scan_sast.pyscript59.1 KB
scripts/scan_secrets.pyscript11.9 KB
scripts/scan_skill_threats.pyscript102.4 KB
scripts/scan_splitstream.pyscript29.5 KB
scripts/scan_yara.pyscript19.7 KB
scripts/session_scan.pyscript54.1 KB
scripts/suppressions.pyscript15.4 KB
scripts/validate_manifests.pyscript13.7 KB
scripts/verify_install.pyscript26.2 KB
scripts/vuln_feed.pyscript31.1 KB
tests/conftest.pyscript13.4 KB
tests/corpus/DEAD_ANCHORS_CORPUS.mdmarkdown2.3 KB
tests/corpus/budgets.jsonconfig5.6 KB
tests/cursor_helpers.pyscript12.6 KB
tests/fixtures/legacy_main_blocks.jsonconfig10.3 KB
tests/fixtures/powershell_verdict_corpus.jsonconfig114.9 KB
tests/golden/parity_mcp_security.jsonconfig17.5 KB
tests/golden/parity_runtime_dynamism.jsonconfig31.7 KB
tests/golden/parity_sast.jsonconfig59.8 KB
tests/golden/parity_secrets.jsonconfig18.8 KB
tests/golden/parity_skill_threats.jsonconfig48.8 KB
tests/parity_corpus.pyscript6.3 KB
tests/powershell_hook.pyscript8.2 KB
tests/report_builders.pyscript5.0 KB
tests/schemas/sarif-schema-2.1.0.jsonconfig109.1 KB
tests/shell_compat.pyscript2.2 KB
tests/test_adjudication_bridge.pyscript832 B
tests/test_adjudication_containment.pyscript1.4 KB
tests/test_adjudication_output.pyscript20.0 KB
tests/test_adjudication_suppressions.pyscript18.7 KB
tests/test_advisory_adjudication.pyscript1.3 KB
tests/test_agent_relay_rules.pyscript7.8 KB
tests/test_aggregate_json.pyscript10.0 KB
tests/test_auto_scan_patterns.pyscript10.5 KB
tests/test_benign_corpus.pyscript16.4 KB
tests/test_benign_new_scanners.pyscript6.1 KB
tests/test_build_output_skip.pyscript10.6 KB
tests/test_capability_ledger.pyscript18.8 KB
tests/test_codex_install.pyscript10.0 KB
tests/test_codex_marketplace.pyscript3.3 KB
tests/test_comment_false_positives.pyscript2.2 KB
tests/test_composer_dependencies.pyscript4.0 KB
tests/test_compromised_versions.pyscript51.2 KB
tests/test_context_gate.pyscript32.4 KB
tests/test_coverage_status.pyscript7.6 KB
tests/test_css_compounding_shrink.pyscript20.8 KB
tests/test_cursor_adapter.pyscript7.8 KB
tests/test_cursor_benign.pyscript3.4 KB
tests/test_cursor_degrade.pyscript12.7 KB
tests/test_cursor_install.pyscript17.1 KB
tests/test_cursor_latency.pyscript7.2 KB
tests/test_cursor_parity.pyscript5.8 KB
tests/test_cursor_redteam_fn.pyscript4.9 KB
tests/test_cursor_tamper_modify.pyscript14.2 KB
tests/test_cursor_wiring.pyscript32.3 KB
tests/test_d1_yara_correlation.pyscript9.2 KB
tests/test_dead_anchors_extract.pyscript7.6 KB
tests/test_dead_anchors_probe.pyscript19.7 KB
tests/test_evidence_engine_parity.pyscript8.1 KB
tests/test_evidence_model.pyscript11.4 KB
tests/test_forensics_core.pyscript79.2 KB
tests/test_fp_hardening_r3.pyscript9.7 KB
tests/test_fp_hardening_r4.pyscript18.8 KB
tests/test_fp_regression_pairs.pyscript62.5 KB
tests/test_fp_structural.pyscript21.2 KB
tests/test_freshness.pyscript6.2 KB
tests/test_frontmatter_validity.pyscript7.0 KB
tests/test_gpg_rce_hotfix.pyscript4.8 KB
tests/test_hook_inert_data.pyscript6.1 KB
tests/test_hook_scripts.pyscript14.0 KB
tests/test_inspection_cpu_budget.pyscript11.6 KB
tests/test_ioc_manager.pyscript55.4 KB
tests/test_kernel_privesc_rules.pyscript27.8 KB
tests/test_legit_precision_regressions.pyscript6.8 KB
tests/test_lethal_trifecta.pyscript20.7 KB
tests/test_manifest_signature.pyscript1.5 KB
tests/test_manifest_validity.pyscript2.9 KB
tests/test_monotonic_trust_e2e.pyscript4.4 KB
tests/test_non_breaking_contract.pyscript22.7 KB
tests/test_openclaw_install.pyscript6.6 KB
tests/test_pack_parity.pyscript24.8 KB
tests/test_parse_pnpm_lock.pyscript10.7 KB
tests/test_plugin_checkout_provenance.pyscript11.1 KB
tests/test_powershell_hook_wire.pyscript10.6 KB
tests/test_powershell_verdict_corpus.pyscript6.2 KB
tests/test_pr44_hardening.pyscript6.5 KB
tests/test_pre_scan.pyscript19.6 KB
tests/test_pre_scan_composer.pyscript7.4 KB
tests/test_precondition_fields.pyscript4.6 KB
tests/test_provenance_offline.pyscript2.5 KB
tests/test_published_rulepack_feed_current.pyscript1.4 KB
tests/test_py38_compat.pyscript3.3 KB
tests/test_pypi_pinned_versions.pyscript19.3 KB
tests/test_pyproject_string_forms.pyscript2.2 KB
tests/test_redteam_fn_regression.pyscript32.5 KB
tests/test_reflective_fetch_execute.pyscript23.5 KB
tests/test_reflective_review_fixes.pyscript17.8 KB
tests/test_refresh_controller.pyscript19.9 KB
tests/test_refresh_rulepacks.pyscript12.4 KB
tests/test_review_gate_regressions.pyscript9.3 KB
tests/test_rule_loader.pyscript23.2 KB
tests/test_rulepack_feed.pyscript38.1 KB
tests/test_run_forensics_json.pyscript1012 B
tests/test_run_scanner_fail_loud.pyscript6.5 KB
tests/test_runtime_floors.pyscript19.8 KB
tests/test_sandbox_probing_rules.pyscript9.6 KB
tests/test_sarif_export.pyscript19.9 KB
tests/test_scan_agent_skills.pyscript31.3 KB
tests/test_scan_archive.pyscript11.6 KB
tests/test_scan_ast.pyscript12.8 KB
tests/test_scan_binary.pyscript15.3 KB
tests/test_scan_bytecode.pyscript21.5 KB
tests/test_scan_dast.pyscript32.1 KB
tests/test_scan_dataflow.pyscript7.1 KB
tests/test_scan_dead_anchors.pyscript11.5 KB
tests/test_scan_decode.pyscript28.0 KB
tests/test_scan_dependencies.pyscript37.8 KB
tests/test_scan_devcontainer.pyscript8.5 KB
tests/test_scan_entropy.pyscript11.2 KB
tests/test_scan_entrypoint.pyscript19.5 KB
tests/test_scan_git_config.pyscript58.5 KB
tests/test_scan_git_config_review_fixes.pyscript33.3 KB

scripts/dead_anchors_extract.py

View on GitHub
#!/usr/bin/env python3
"""
dead_anchors_extract.py - Anchor extraction for the dead-anchor / repojacking
scanner (U1).

Given a repo path, pulls every EXTERNAL anchor a skill/repo points at, with ZERO
network calls:
  - GitHub owner/repo (URL form DA-GH-001, `github:` shorthand DA-GH-002),
    reserved-word owners (DA-RW-001) filtered out.
  - Prose package-install targets (DA-PK-001); pip->PyPI, npm/yarn/pnpm->npm;
    gem/cargo/bundle are extracted but tagged ecosystem=None (no P0 probe).
  - Bare domains (DA-DM-001) reduced to eTLD+1 via the vendored multi-part-TLD
    list (DA-TLD-001); safe-domain allowlist (DA-SD-001) skipped.
  - Free-tier cloud subdomains (DA-CL-001 suffix membership).

Everything data-driven lives in data/rulepacks/dead_anchors.json (feed-updatable,
read-only at scan time). Anchors are DE-DUPLICATED before return so N references
to the same target cost exactly one probe downstream (loop-safety, spec §10).

Never raises on a scan target: a missing/invalid pack => "nothing to probe this
run" (empty list), a malformed seed file => skipped, never a crash.

Created by Alex Greenshpun
"""

import os
import re
import sys
import stat
from collections import namedtuple

_O_NONBLOCK = getattr(os, "O_NONBLOCK", 0)

_SCRIPTS_DIR = os.path.dirname(os.path.abspath(__file__))
if _SCRIPTS_DIR not in sys.path:
    sys.path.insert(0, _SCRIPTS_DIR)

import rule_loader

try:
    from _shared_patterns import SEED_FILES as _SEED_FILES
except Exception:  # pragma: no cover - fallback if shared pack unavailable
    _SEED_FILES = {
        'SKILL.md', 'SOUL.md', 'HEARTBEAT.md', 'ROUTINE.md', 'AGENTS.md',
        'BOOT.md', 'BOOTSTRAP.md', 'CLAUDE.md', 'IDENTITY.md', 'USER.md',
    }

# Additional non-seed manifest files worth scanning for anchors.
_EXTRA_FILES = {".mcp.json", "mcp.json", "README.md", "readme.md"}

# Directories we never descend into (heavy / vendored / vcs).
_SKIP_DIRS = {
    ".git", "node_modules", ".venv", "venv", "__pycache__", "dist", "build",
    ".tox", ".mypy_cache", ".pytest_cache", "site-packages",
}

# Bounds (loop-safety / lightweight): never read an unbounded tree.
_MAX_FILES = 200
_MAX_FILE_BYTES = 1_000_000
# Hard cap on distinct extracted anchors so a maximally-adversarial (but
# within-file-count/size-bound) repo cannot drive extraction / the downstream
# dispatch loop into multi-minute / multi-GB territory (python-redos F3).
_MAX_ANCHORS = 5000


def normalize_npm_name(name):
    """npm package identity is case-insensitive and lowercase; an @scope is kept
    but lowercased too. Used for BOTH the existence probe and the dedup key so
    'npm install React' and 'react' resolve to one live package, not a phantom."""
    return (name or "").strip().lower()


def normalize_pypi_name(name):
    """PEP 503 canonical form: lowercase and collapse any run of -, _ or . to a
    single -. 'My_Package', 'my-package', 'my.package' are one distribution."""
    return re.sub(r"[-_.]+", "-", (name or "").strip()).lower()

# Verbs near a domain that promote DA-04 from HIGH to CRITICAL (the domain is an
# active fetch/install target, not a passing mention). Best-effort, per spec §2.
_FETCH_VERB_RE = re.compile(
    r"(?i)\b(curl|wget|fetch|download|source|install|clone|git\s+clone|pip\s+install|"
    r"npm\s+install|require|import\s+from)\b"
)

# One immutable anchor record. `raw` is the original reference (for the snippet);
# `target` is the normalized probe key; `is_free_tier` flags DA-09.
Anchor = namedtuple(
    "Anchor",
    ["type", "target", "owner", "repo", "ecosystem", "suffix",
     "is_free_tier", "file", "line", "raw", "fetch_context"],
)


def _rule_by_id(pack, rule_id):
    if pack is None:
        return None
    try:
        rules = pack.all_rules
    except Exception:
        return None
    for r in rules:
        if getattr(r, "id", None) == rule_id:
            return r
    return None


def _keyword_values(rule):
    """Values of a keyword rule, or an empty set for anything else. A pack that
    has been tampered/corrupted so this id is authored as a NON-keyword type
    (rule.values is then None) must degrade to 'no data', never crash on
    set(None) (python-redos F2 / error-soundness never-hard-fail)."""
    if rule is None or getattr(rule, "type", None) != "keyword":
        return set()
    vals = getattr(rule, "values", None)
    if not vals:
        return set()
    try:
        return set(vals)
    except TypeError:
        return set()


class _Extractor:
    """Compiled pack view. Built once per scan; holds the four extraction
    regexes and the four static lists. A missing pack degrades to a no-op
    extractor (every extract_* returns nothing) rather than crashing."""

    def __init__(self, pack=None, base_dir=None):
        if pack is None:
            try:
                pack = rule_loader.load_pack("dead_anchors", base_dir=base_dir)
            except Exception:
                pack = None
        self.pack = pack
        gh = _rule_by_id(pack, "DA-GH-001")
        sh = _rule_by_id(pack, "DA-GH-002")
        pk = _rule_by_id(pack, "DA-PK-001")
        dm = _rule_by_id(pack, "DA-DM-001")
        self.re_gh = getattr(gh, "regex", None) if gh else None
        self.re_sh = getattr(sh, "regex", None) if sh else None
        self.re_pk = getattr(pk, "regex", None) if pk else None
        self.re_dm = getattr(dm, "regex", None) if dm else None
        cl = _rule_by_id(pack, "DA-CL-001")
        rw = _rule_by_id(pack, "DA-RW-001")
        sd = _rule_by_id(pack, "DA-SD-001")
        tld = _rule_by_id(pack, "DA-TLD-001")
        # rule.values are NFKC-lowercased tuples; domains/suffixes are already
        # lowercase so this is lossless. _keyword_values tolerates a tampered
        # pack that authors any of these ids as a non-keyword type.
        self.cloud_suffixes = _keyword_values(cl)
        self.reserved_words = _keyword_values(rw)
        self.safe_domains = _keyword_values(sd)
        self.multi_tlds = _keyword_values(tld)

    # -- individual extractors (operate on one line of text) ------------------

    def _github_from(self, regex, line, ecosystem_hint):
        out = []
        if regex is None:
            return out
        for m in regex.finditer(line):
            owner, repo = m.group(1), m.group(2)
            if not owner or not repo:
                continue
            if owner.lower() in self.reserved_words:
                continue
            # Trailing punctuation cleanup (e.g. "repo)." in prose).
            repo = repo.rstrip(".")
            if not repo:
                continue
            out.append((owner, repo, m.group(0)))
        return out

    def _packages_from(self, line):
        out = []
        if self.re_pk is None:
            return out
        for m in self.re_pk.finditer(line):
            verb = m.group(0).split()[0].lower()
            raw_name = m.group(1)
            name = _strip_pkg_version(raw_name)
            if not name:
                continue
            if verb.startswith("pip"):
                eco = "PyPI"
            elif verb in ("npm", "yarn", "pnpm"):
                eco = "npm"
            else:
                eco = None  # gem/cargo/bundle: extracted, not probed in P0
            out.append((name, eco, m.group(0)))
        return out

    def _domains_from(self, line):
        """Yield (kind, payload, raw) where kind is 'cloud' or 'domain'."""
        out = []
        if self.re_dm is None:
            return out
        for m in self.re_dm.finditer(line):
            host = m.group(1).lower().rstrip(".")
            if not host or "." not in host:
                continue
            suffix = self._cloud_suffix_for(host)
            if suffix is not None:
                if host == suffix:
                    continue  # bare platform domain, not our subdomain to claim
                out.append(("cloud", (host, suffix), m.group(0)))
                continue
            reg = self._registrable(host)
            if not reg or reg in self.safe_domains:
                continue
            out.append(("domain", reg, m.group(0)))
        return out

    def _cloud_suffix_for(self, host):
        for suf in self.cloud_suffixes:
            if host == suf or host.endswith("." + suf):
                return suf
        return None

    def _registrable(self, host):
        labels = host.split(".")
        if len(labels) < 2:
            return None
        # Longest matching multi-part TLD wins.
        for n in (3, 2):
            if len(labels) > n:
                candidate = ".".join(labels[-n:])
                if candidate in self.multi_tlds:
                    return ".".join(labels[-(n + 1):])
        return ".".join(labels[-2:])


def _strip_pkg_version(raw):
    """Strip version specifiers/flags from a captured package token, preserving
    an npm scope (@scope/name)."""
    name = raw.strip().strip(",;)")
    if not name:
        return ""
    # pip-style specifiers.
    for sep in ("==", ">=", "<=", "~=", "!=", ">", "<", "["):
        idx = name.find(sep)
        if idx > 0:
            name = name[:idx]
    # npm @version, keeping a leading scope @.
    if name.startswith("@"):
        parts = name[1:].split("@", 1)
        name = "@" + parts[0]
    else:
        name = name.split("@", 1)[0]
    return name.rstrip("/.")


def _iter_seed_files(repo_path):
    """Yield (abs_path, rel_path) for seed / manifest files, bounded.

    Non-regular files (FIFOs, sockets, device nodes) are skipped BEFORE they are
    ever handed to a reader: a FIFO named 'SKILL.md' shipped in a malicious repo
    would otherwise block open()/read() forever and hang the whole scan
    (python-redos F1). os.stat follows a symlink to its target, so a symlink
    pointing at a FIFO/device is caught here too; only a real regular file (of
    the target) passes."""
    seeds_lower = {s.lower() for s in _SEED_FILES} | {e.lower() for e in _EXTRA_FILES}
    count = 0
    for root, dirs, files in os.walk(repo_path):
        dirs[:] = [d for d in dirs if d not in _SKIP_DIRS]
        for fn in files:
            if fn.lower() in seeds_lower:
                abs_path = os.path.join(root, fn)
                try:
                    st = os.stat(abs_path)  # follows symlink to its final target
                except OSError:
                    continue
                if not stat.S_ISREG(st.st_mode):
                    continue  # FIFO / socket / device / dir -> never open
                if st.st_size > _MAX_FILE_BYTES:
                    continue  # size cap enforced BEFORE any read
                rel = os.path.relpath(abs_path, repo_path)
                yield abs_path, rel
                count += 1
                if count >= _MAX_FILES:
                    return


def _read_regular_file(abs_path):
    """Open a seed file defensively and return its text, or None to skip it.

    Belt-and-suspenders over _iter_seed_files: O_NONBLOCK guarantees the open()
    itself can never block on a FIFO/device even under a TOCTOU swap after the
    stat, and an fstat on the fd we actually hold re-confirms it is a regular
    file and re-checks the size cap on that exact fd."""
    try:
        fd = os.open(abs_path, os.O_RDONLY | _O_NONBLOCK)
    except OSError:
        return None
    try:
        st = os.fstat(fd)
        if not stat.S_ISREG(st.st_mode) or st.st_size > _MAX_FILE_BYTES:
            os.close(fd)
            return None
    except OSError:
        try:
            os.close(fd)
        except OSError:
            pass
        return None
    try:
        with os.fdopen(fd, "r", encoding="utf-8", errors="replace") as f:
            return f.read()
    except (OSError, ValueError):
        try:
            os.close(fd)
        except OSError:
            pass
        return None


def extract_anchors(repo_path, pack=None, base_dir=None):
    """Extract the deduped list of Anchor records from a repo. Zero network.

    De-dup key is (type, target) so N references to the same anchor => 1 probe.
    The FIRST occurrence's file/line/raw is kept for the finding location.
    """
    try:
        ex = _Extractor(pack=pack, base_dir=base_dir)
    except Exception:
        return []  # tampered/malformed pack: nothing to probe this run (no crash)
    if not (ex.re_gh or ex.re_sh or ex.re_pk or ex.re_dm):
        return []  # pack unavailable: nothing to probe this run

    seen = {}  # normalized (type, key) -> Anchor (first wins)

    def _add(anchor):
        # Dedup on the NORMALIZED identity so case/separator variants of the same
        # real anchor collapse to one probe + one finding. The original casing is
        # preserved on the stored Anchor for display.
        if anchor.type == "github":
            key = ("github", "%s/%s" % ((anchor.owner or "").lower(),
                                        (anchor.repo or "").lower()))
        elif anchor.type == "package":
            eco, _, nm = anchor.target.partition(":")
            if eco == "npm":
                nm = normalize_npm_name(nm)
            elif eco == "PyPI":
                nm = normalize_pypi_name(nm)
            key = ("package", "%s:%s" % (eco, nm))
        else:
            key = (anchor.type, anchor.target)
        if key not in seen:
            seen[key] = anchor

    for abs_path, rel in _iter_seed_files(repo_path):
        if len(seen) >= _MAX_ANCHORS:
            break
        text = _read_regular_file(abs_path)
        if text is None:
            continue
        lines = text.splitlines(keepends=True)
        for lineno, line in enumerate(lines, 1):
            if len(seen) >= _MAX_ANCHORS:
                break
            if len(line) > 8000:
                line = line[:8000]
            fetch_ctx = bool(_FETCH_VERB_RE.search(line))
            # GitHub URL + shorthand.
            for owner, repo, raw in ex._github_from(ex.re_gh, line, None):
                _add(Anchor("github", f"{owner}/{repo}", owner, repo, None,
                            None, False, rel, lineno, raw, fetch_ctx))
            for owner, repo, raw in ex._github_from(ex.re_sh, line, None):
                _add(Anchor("github", f"{owner}/{repo}", owner, repo, None,
                            None, False, rel, lineno, raw, fetch_ctx))
            # Packages.
            for name, eco, raw in ex._packages_from(line):
                if eco is None:
                    continue  # gem/cargo/bundle not probed in P0
                _add(Anchor("package", f"{eco}:{name}", None, None, eco,
                            None, False, rel, lineno, raw, fetch_ctx))
            # Domains + cloud subdomains.
            for kind, payload, raw in ex._domains_from(line):
                if kind == "cloud":
                    host, suffix = payload
                    _add(Anchor("cloud", host, None, None, None, suffix,
                                True, rel, lineno, raw, fetch_ctx))
                else:
                    _add(Anchor("domain", payload, None, None, None, None,
                                False, rel, lineno, raw, fetch_ctx))

    return list(seen.values())
rea-analysis-optimizationmorluto/reaOptimize REA providers and shared analysis workflows for agent context, latency, CPU, and memory while preserving correctness and Evidence. Use for analysis bottlenecks and cache or query design; CI/test-suite optimization belongs to rea-ci-tests-audit.Stars: 472kUpdated Claude CodeCodexVerifiedInstall
rea-ci-tests-auditmorluto/reaAudit REA CI and test suites for correctness, fixture drift, false passes, and measured performance; use for suite or workflow investigations and efficiency changes, rather than ordinary feature testing or a single test failure.Stars: 472kUpdated Claude CodeCodexVerifiedInstall
rea-tool-designmorluto/reaAudit, investigate, design, or change REA investigation tools, CLI/MCP contracts, provider capabilities, and Evidence semantics.Stars: 472kUpdated Claude CodeCodexVerifiedInstall
ponytail-auditDietrichGebert/ponytailQuality audit of the whole repo: bugs, security, real load, missing tests, speed, and what to delete. Most important first.Stars: 194kUpdated Claude CodeCodexVerifiedInstall