Skip to content

j-mauro-r/SAT-Dengue-V2 / .agents/skills/deepworkplan

deepworkplan

Verified

DeepWorkPlan — turn any repo AI-first and run Deep Work Plans. Routes to create, execute, refine, resume, status, verify, upgrade, and repo-onboarding sub-skills based on intent. Use when the developer wants to plan, execute, manage, or verify structured multi-task work, or make a repository AI-agent-ready.

FileKindSize
SKILL.mdskill md12.0 KB
TRUST.mdmarkdown16.1 KB
addons/README.mdmarkdown11.7 KB
addons/machine-installs.mdmarkdown2.9 KB
examples/CREATE_PLAN.mdmarkdown6.9 KB
examples/ORCHESTRATOR_TASK_TEMPLATE_create_child_dwp.mdmarkdown15.0 KB
examples/ORCHESTRATOR_TASK_TEMPLATE_execute_child_dwp.mdmarkdown16.7 KB
examples/ORCHESTRATOR_TASK_TEMPLATE_integration_checkpoint.mdmarkdown9.0 KB
examples/PROMPTS_TEMPLATE.mdmarkdown4.0 KB
examples/TEAM_AGENTS_TASK_TEMPLATE.mdmarkdown5.1 KB
examples/V6_PLAN_EXAMPLE.mdmarkdown5.5 KB
guide/GUIDE.mdmarkdown8.0 KB
guide/authoring.mdmarkdown28.5 KB
guide/execution.mdmarkdown21.0 KB
guide/large-repo-onboarding.mdmarkdown3.3 KB
guide/orchestration-at-scale.mdmarkdown11.4 KB
guide/orchestrator.mdmarkdown37.9 KB
guide/prompts.mdmarkdown10.1 KB
guide/skills-integration.mdmarkdown6.5 KB
guide/structure.mdmarkdown6.2 KB
guide/swarm.mdmarkdown21.2 KB
guide/team-agents.mdmarkdown12.3 KB
shared/adaptation.mdmarkdown2.7 KB
shared/addon_sync.pyscript17.7 KB
shared/benchmark.pyscript97.6 KB
shared/config.pyscript41.5 KB
shared/context.shscript4.2 KB
shared/context_manifest.pyscript40.8 KB
shared/contract_v6.pyscript82.7 KB
shared/delegators.pyscript3.0 KB
shared/dwp-paths.mdmarkdown5.8 KB
shared/finalize_plan.pyscript7.2 KB
shared/install-verification.mdmarkdown1.8 KB
shared/ledger.pyscript141.5 KB
shared/migrate_v6.pyscript46.5 KB
shared/outcomes.pyscript27.2 KB
shared/plan_paths.pyscript4.8 KB
shared/resources.pyscript49.0 KB
shared/scheduler.pyscript41.6 KB
shared/state_contract.pyscript9.8 KB
shared/swarm.pyscript27.3 KB
shared/troubleshooting.mdmarkdown7.4 KB
shared/update-state.pyscript12.5 KB
shared/views.pyscript21.0 KB
shared/working-principles.mdmarkdown7.1 KB
spec/ADDONS.mdmarkdown27.4 KB
spec/AGENT_PROTOCOL.mdmarkdown28.0 KB
spec/ARCHETYPES.mdmarkdown10.6 KB
spec/BENCHMARK.mdmarkdown20.3 KB
spec/CONFIG.mdmarkdown8.7 KB
spec/DOCUMENTATION_STANDARD.mdmarkdown40.7 KB
spec/DWP_SPECIFICATION.mdmarkdown65.1 KB
spec/LITE_PLANS.mdmarkdown4.0 KB
spec/PLAN_STATE.mdmarkdown34.9 KB
spec/README.mdmarkdown11.5 KB
spec/V6_CONTEXT.mdmarkdown6.4 KB
spec/V6_CONTRACT.mdmarkdown12.3 KB
spec/V6_LIFECYCLE.mdmarkdown13.5 KB
spec/V6_RESOURCES.mdmarkdown4.7 KB
spec/V7_ABILITIES.mdmarkdown4.9 KB
spec/V7_CONTRACT.mdmarkdown6.4 KB
spec/V7_ROADMAP.mdmarkdown4.6 KB
spec/schema/addon-descriptor-v1.schema.jsonconfig5.5 KB
spec/schema/benchmark-record.schema.jsonconfig14.0 KB
spec/schema/context-manifest-v6.schema.jsonconfig12.5 KB
spec/schema/dwp-config-v1.schema.jsonconfig3.5 KB
spec/schema/dwp-handoff-v1.schema.jsonconfig3.0 KB
spec/schema/journal-event-v6.schema.jsonconfig22.3 KB
spec/schema/journal-event-v7.schema.jsonconfig24.9 KB
spec/schema/learnings-record.schema.jsonconfig6.0 KB
spec/schema/plan-contract-v6.schema.jsonconfig16.4 KB
spec/schema/plan-contract-v7.schema.jsonconfig16.6 KB
spec/schema/plan-manifest-v2.schema.jsonconfig1.0 KB
spec/schema/plan-manifest-v5.schema.jsonconfig1.3 KB
spec/schema/plan-manifest-v6.schema.jsonconfig2.0 KB
spec/schema/plan-manifest-v7.schema.jsonconfig1.7 KB
spec/schema/plan-manifest.schema.jsonconfig1.8 KB
spec/schema/plan-snapshot-v6.schema.jsonconfig7.8 KB
spec/schema/plan-state-v2.schema.jsonconfig6.4 KB
spec/schema/plan-state-v5.schema.jsonconfig6.7 KB
spec/schema/plan-state.schema.jsonconfig3.7 KB

spec/CONFIG.md

View on GitHub
# CONFIG.md — DeepWorkPlan configuration file (Normative)

> Status: **7.0.0**. Defines `.dwp/config.json` and
> `~/.dwp/config.json`: one file shape, one parser
> ([`../shared/config.py`](../shared/config.py)), two keys — `benchmark`
> ([`BENCHMARK.md`](BENCHMARK.md) §1) and `addons` (the addon registry,
> §3 below). Published as
> [`schema/dwp-config-v1.schema.json`](schema/dwp-config-v1.schema.json)
> (`https://deepworkplan.com/schema/dwp-config/v1.json`). RFC-2119 language.

**Governing principle.** The methodology works with no configuration file
and no addon. Nothing in this file can make a plan, a flow or conformance
depend on an addon: the registry records what a repository *opted into*, so
flows may **offer** or **amplify** — never require.

## 1. Locations and precedence

| Precedence | Path | Scope |
|---|---|---|
| 1 (highest) | `<repo-root>/.dwp/config.json` | the repository that owns the plan (the plan's `.dwp` ancestor), or the repository a flow runs in when there is no plan |
| 2 | `~/.dwp/config.json` | every repository of the local user |
| 3 | absent | everything disabled |

**The repository file is tracked** so teammates and CI read the same
registry (F-04): onboarding writes the `.gitignore` rule `.dwp/*` +
`!.dwp/config.json` — plans, onboarding notes and benchmark records stay
ignored. A repository with a plain `.dwp/` rule stays conformant; its
registry is then local. The file holds decisions, never secrets.

Both files share one shape:

```json
{
  "benchmark": { "enabled": true, "learnings": true },
  "addons": {
    "ai-diff-reviewer": { "enabled": true, "version": "v3.3.14" },
    "herdr":            { "enabled": true, "version": "v1.0.0" },
    "agentkit":         { "enabled": false },
    "devcontainer":     { "enabled": true, "version": "v1.0.1" },
    "vim":              { "enabled": true, "version": "v1.0.0" },
    "dailybot":         { "enabled": true, "version": "v3.23.17" }
  }
}
```

- An optional top-level `"schema"` MAY name
  `https://deepworkplan.com/schema/dwp-config/v1.json`; readers do not
  require it. Unknown top-level keys are ignored (forward compatibility).
- **Resolution differs by key.** `benchmark` resolves **wholesale**: a
  repository file that carries the object overrides the user file for both
  of its flags (`BENCHMARK.md` §1). `addons` resolves **per addon key**:
  the repository entry for a key wins over the user entry for that key; a
  repository file that omits a key defers to the user file for it.

## 2. Fail-closed reading (both keys)

- A missing file is the ordinary "omits everything" case — no warning.
- An unreadable path or invalid JSON: the file contributes nothing, with
  exactly **one** warning naming the file and the reason; the other file
  still applies.
- A wrong-typed value disables only what it affects, with **one** warning
  naming the file, the key and the reason (`benchmark` keeps its §1 rules;
  `addons` per §3).
- A reader MUST NOT raise, MUST NOT abort a flow, and MUST NOT be silent
  about a value it discarded.

## 3. The addon registry (`addons`)

- **Key set.** The registry keys are the **in-pack addon directory names**
  (`addons/<key>/`; ecosystem contract amendment A1) — today
  `ai-diff-reviewer`, `agentkit`, `dailybot`, `dependency-upgrade`,
  `design-system`, `devcontainer`, `herdr`, `vim`. Each addon ships an
  `addon.json` descriptor whose `key` equals its directory name
  ([`ADDONS.md`](ADDONS.md) §7). The reader derives the set by listing
  directories; it never opens a file inside an addon to do so. A key
  outside the set is **ignored with one warning** (a newer pack's addon, a
  typo) — never an error.
- **Entry shape (closed).** `enabled` (boolean, **required to enable**;
  only `true` enables), optional `version` (string matching
  `^v\d+\.\d+\.\d+(-[0-9A-Za-z.]+)?$`) and optional `note` (one line,
  1–200 characters — e.g. "accepted; machine-level install deferred").
  An enabled addon that is not detected and carries a note is reported as
  **deferred** (`resources.py abilities` → `deferred`), without a warning
  (F-05). Any other field, a missing
  `enabled`, a non-boolean `enabled` or a malformed `version` makes that
  entry **not enabled** with one warning naming the file, key and reason.
  A malformed entry in the repository file decides its key (fail-closed);
  it does not fall through to the user file.
- **`"addons"` itself** must be an object; anything else contributes no
  entry from that file, with one warning.
- **Absent = not enabled.** Nothing is enabled by default. Only what
  onboarding recorded with consent, an addon's own install step, or the
  human enables anything.
- **The AI Diff Reviewer baseline is unchanged.** Its local review stays
  part of the Final Review (`ADDONS.md` §6.5); a missing installation is the
  recorded `local reviewer not installed` finding whether or not the
  registry names it. The registry adds no requirement.
- **`version` is informative in 7.0.0.** `verify` MAY warn when it differs
  from the installed product's observed version; it MUST NOT fail on it.
- **Enabled is necessary, not sufficient.** An addon contributes abilities
  only when it is enabled **and** its descriptor's detection succeeds with
  a compatible interface major (`V7_ABILITIES.md`); being listed never
  makes a product present.

## 3a. The host capability record (`host`)

The abilities of the host that runs plans here, machine-readable (F-17):
`"host": {"subagents": true, "cancel_children": true}` — the closed v6 set
(`stop_agent`, `meter_spend`, `meter_tokens`, `meter_wall_clock`,
`cancel_children`, `model_routing`, `subagents`, `telemetry`), booleans
only. Per capability the user file wins over the repository file (the
record describes a machine; the tracked file is the team baseline), and an
explicit `resources.py --caps` declaration wins over both; an unstated
capability stays at the all-False floor. An unknown or non-boolean
capability is ignored with one warning — never invented. `resources.py
abilities` reads it and names where each declared capability came from
(`host_declared`). Writer: `config.py host <capability> true|false --repo
<repo>`, run by onboarding when it records the host declaration
([`../onboard/v6.md`](../onboard/v6.md) §1); the prose block in
`AGENTS.md` stays the human summary. `telemetry: true` still needs the
developer's consent; the record states the host can, not that it may.

## 4. Writers and readers

| Role | Who | How |
|---|---|---|
| Writer | `onboard` Phase 7b, on the developer's acceptance of one addon's offer | `python3 <pack>/shared/config.py enable <key> [--version <tag>] --repo <repo>` (a decline writes nothing; an explicit "turn it off" is `disable <key>`) |
| Writer | an addon's own install step, or the human | the same command, or a hand edit |
| Writer | the developer, or an agent they asked, turning field metrics on | `python3 <pack>/shared/config.py benchmark enabled\|learnings true\|false --repo <repo>` (7.1.5): sets one `benchmark` key, preserves the other and every other top-level key, refuses an unknown key or a file it cannot parse. `learnings` still rides on `enabled` (BENCHMARK.md §1). `enable`/`disable` stay addon-only |
| Writer | `upgrade`, after the developer accepts the upgrade | `config.py backfill --repo <repo>` (dry run), then `--write`: every in-pack addon the repository file does not name and that is installed **in the repository** (a repo-relative detect path) is recorded `enabled` with its observed version and a back-fill note (F-15); a machine-level install is reported, never recorded — onboarding offers it. A recorded decision — enabled or disabled — is never changed |
| Reader | `execute`, `create`, `verify`, `status` | `python3 <pack>/shared/config.py show` / `enabled` (`--plan <dir>` resolves the plan's repository) |
| Reader | `shared/benchmark.py` | the `benchmark` key only, at its emission point |

The writer **reconciles**: it preserves every other top-level key and every
other addon entry, writes atomically, and **refuses** (exit 2, nothing
written) an unknown key, a malformed version, or an existing file it cannot
parse — it never overwrites what it cannot read.

The `show` view is `{"benchmark": {...}, "addons": {key: {"enabled",
"version", "source"}}}` with warnings on stderr; `enabled` prints the
enabled keys one per line. Both are read-only.

## 5. Conformance

- A repository with no configuration file is fully conformant, and every
  flow behaves identically with no file, with every addon disabled, or with
  only unknown keys (`tests/standalone-methodology.bats`).
- The configuration file is never a plan artifact: plans never persist the
  registry or the abilities it contributes (`V7_ABILITIES.md`); a plan
  records only what it **used**, in its journal.
reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis.Stars: 472kUpdated Claude CodeCodexVerifiedInstall
diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness.Stars: 322kUpdated Claude CodeCodexVerifiedInstall
dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependenciesStars: 322kUpdated Claude CodeCodexVerifiedInstall
executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is availableStars: 322kUpdated Claude CodeCodexVerifiedInstall