Skip to content

BdM-15/capture-insights / skills/competitive-intel

competitive-intel

Verified

Federal capture competitor, incumbent, and obligation intelligence backed by live USAspending.gov data via the vendored `usaspending` MCP. USE WHEN the user asks "who's the incumbent on this contract?", "what was the last award value?", "give me a black-hat read on Bidder X", "who are the top three competitors for this NAICS?", "what's typical pricing for this work?", "pull award history for this agency / NAICS / PSC", "analyze burn rate", "obligation trend by year", or "start from this contract number". Supports both black-hat competitor research and contract-number-first obligation analysis, including standalone awards, parent IDIQs, and orders under IDIQs with parent/child rollups. DO NOT USE FOR proposal drafting (use `proposal-generator`), clause / FAR compliance audit (use `compliance-auditor`), or generic web search about a company (no MCP tool covers that yet).

FileKindSize
SKILL.mdskill md29.0 KB
evals/evals.jsonconfig11.7 KB
references/black_hat_template.mdmarkdown1.5 KB
references/data_sources.mdmarkdown10.2 KB

references/black_hat_template.md

View on GitHub
# Black-Hat One-Pager — canonical structure (Phase 4d)

Use this template once per competitor (incumbent + top 3 likely). Every
non-trivial claim cites either an `award_id` (USAspending
`generated_internal_id`) or a `chunk_id` (Theseus KG). Anything you cannot
cite goes in `claim_gaps`, NOT in the body.

```
COMPETITOR: <recipient name>
ROLE: incumbent | likely-1 | likely-2 | likely-3
RECIPIENT_HASH: <from get_recipient_profile>
PARENT_RECIPIENT: <parent name or "self">
BUSINESS_TYPES: <e.g. "Large Business, For-Profit Organization">

LAST 3 WINS IN THIS NAICS / AGENCY:
  1. <agency> · $<amount> · <pop_end> · [award:<id>]
  2. ...
  3. ...

PREDICTED THEMES (anchored to award history):
  1. <verb-led theme> — evidence: [award:<id>], [award:<id>]
  2. <verb-led theme> — evidence: [award:<id>]
  3. ...

CLAIM GAPS (what we cannot verify with USAspending alone):
  - CPARS performance ratings — no MCP source yet
  - Protest history — no MCP source yet
  - Recent news / leadership changes — no MCP source yet
  - Teaming intent on this specific opportunity — qualitative, requires capture-team intel

GHOST LANGUAGE HOOKS (for proposal-generator):
  - <hook 1 — short, evaluator-friendly phrase>
  - <hook 2>

NOTES FOR USER (manual research):
  - <thing the user should look up before red team>
```

Fill ONLY the fields you have evidence for. Drop the line entirely if
empty rather than emit a "TBD" — `proposal-generator` consumes this and
treats placeholder strings as content.
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