$reverse-engineer-rpi

Reverse-engineer a product into a mechanically verifiable feature inventory + registry + spec set, with optional security-audit artifacts and validation gates.

Hard Guardrails (MANDATORY)

• Only operate on code/binaries you own or have explicit written authorization to analyze.
• Do not provide steps to bypass protections/ToS or to extract proprietary source code/system prompts from third-party products.
• Do not output reconstructed proprietary source or embedded prompts from binaries (index only; redact in reports).
• Redact secrets/tokens/keys if encountered; run the secret-scan gate over outputs.
• Always separate: docs say vs code proves vs hosted/control-plane.

One-Command Example

python3 skills/reverse-engineer-rpi/scripts/reverse_engineer_rpi.py ao \
  --authorized \
  --mode=binary \
  --binary-path="$(command -v ao)" \
  --output-dir=".agents/research/ao/"

If you do not have explicit written authorization to analyze that binary, do not run the above. Use the included demo fixture instead (see Self-Test below).

Repo-only example (no binary required):

python3 skills/reverse-engineer-rpi/scripts/reverse_engineer_rpi.py cc-sdd \
  --mode=repo \
  --upstream-repo="https://github.com/gotalab/cc-sdd.git" \
  --output-dir=".agents/research/cc-sdd/"

Pinned clone (reproducible):

python3 skills/reverse-engineer-rpi/scripts/reverse_engineer_rpi.py cc-sdd \
  --mode=repo \
  --upstream-repo="https://github.com/gotalab/cc-sdd.git" \
  --upstream-ref=v1.0.0 \
  --output-dir=".agents/research/cc-sdd/"

Invocation Contract

Required:

• product_name

Optional:

• --docs-sitemap-url (recommended when available; supports https://... and file:///...)
• --docs-features-prefix (default: auto; detects best local docs prefix, falls back to docs/features/)
• --upstream-repo (optional)
• --upstream-ref (pin clone to a specific commit, tag, or branch; records resolved SHA in clone-metadata.json)
• --local-clone-dir (default: .tmp/<product_name>)
• --output-dir (default: .agents/research/<product_name>/)
• --mode (default: repo; allowed: repo|binary|both)
• --binary-path (required if --mode includes binary)
• --no-materialize-archives (authorized-only; binary mode extracts embedded ZIPs by default; this disables extraction and keeps index-only)

Security audit flags (optional):

• --security-audit (enables security artifacts + gates)
• --sbom (generate SBOM + dependency risk report where possible; may no-op with a note)
• --fuzz (only if a safe harness exists; timeboxed)

Mandatory guardrail flag:

• --authorized (required for binary mode; refuses to run binary analysis without it)

Upstream Ref Pinning (--upstream-ref)

Use --upstream-ref to pin a repo-mode clone to a specific commit, tag, or branch. This makes analysis reproducible and allows golden fixtures to be diffed against a known baseline.

# Pin to a tag (reproducible)
python3 skills/reverse-engineer-rpi/scripts/reverse_engineer_rpi.py cc-sdd \
  --mode=repo \
  --upstream-repo="https://github.com/gotalab/cc-sdd.git" \
  --upstream-ref=v1.0.0 \
  --output-dir=".agents/research/cc-sdd/"

# Pin to a specific commit SHA
python3 skills/reverse-engineer-rpi/scripts/reverse_engineer_rpi.py cc-sdd \
  --mode=repo \
  --upstream-repo="https://github.com/gotalab/cc-sdd.git" \
  --upstream-ref=abc1234 \
  --output-dir=".agents/research/cc-sdd/"

When --upstream-ref is provided:

• The clone is fetched with git fetch --depth=1 origin <ref> and checked out to FETCH_HEAD.
• The resolved commit SHA is recorded in output_dir/clone-metadata.json for traceability.
• Without --upstream-ref, a --depth=1 shallow clone of the default branch HEAD is used instead.

clone-metadata.json schema:

{
  "upstream_repo": "https://github.com/gotalab/cc-sdd.git",
  "upstream_ref": "v1.0.0",
  "resolved_commit": "<full SHA>",
  "clone_date": "YYYY-MM-DD"
}

Contract Outputs (output_dir/)

Repo-mode analysis writes machine-checkable contract files under output_dir/. These files use only relative paths, sorted lists, and stable keys — no absolute paths, no run-specific timestamps — so they can be committed as golden fixtures and diffed across runs.

Primary contract files:

Example feature-registry.yaml structure:

schema_version: 1
product_name: cc-sdd
upstream_commit: "abc1234..."
features:
  - name: cli-entry
    cli:
      language: node
      bin:
        cc-sdd: dist/cli.js
      help_text: "Usage: cc-sdd [options] ..."
  - name: config-surface
    config_env:
      config_file: ".cc-sdd/config.json"
      env_vars:
        - name: CC_SDD_TOKEN
          evidence: ["src/config.ts"]

Note: Contract outputs are written by --mode=repo (or --mode=both). Binary-mode outputs (binary-analysis.md, binary-symbols.txt, etc.) remain directly under output_dir/.

Fixture Test Workflow

Golden fixtures allow regression detection: commit a known-good fixture snapshot (contract files alongside the pinned clone-metadata.json), then diff future runs against it.

Running Fixture Tests

bash skills/reverse-engineer-rpi/scripts/repo_fixture_test.sh

This script (implemented in ag-w77.3):

1. Reads skills/reverse-engineer-rpi/fixtures/cc-sdd-v2.1.0/clone-metadata.json to determine the pinned upstream ref.
2. Runs reverse_engineer_rpi.py in repo mode with that ref into a temp output dir.
3. Diffs the generated outputs against the committed golden fixtures (feature-registry.yaml, cli-surface-contracts.txt, docs-features.txt).
4. Exits 0 if they match; exits non-zero with a unified diff if they drift.

The test requires network access to clone the upstream repo.

Updating Fixtures

When contracts legitimately change (new flags, new env vars, schema bumps), update the golden fixtures:

# 1. Re-run with the pinned ref to generate fresh contracts
python3 skills/reverse-engineer-rpi/scripts/reverse_engineer_rpi.py cc-sdd \
  --mode=repo \
  --upstream-repo="https://github.com/gotalab/cc-sdd.git" \
  --upstream-ref=<new-tag-or-sha> \
  --output-dir=".tmp/cc-sdd-refresh/"

# 2. Copy contracts into the fixture directory
cp .tmp/cc-sdd-refresh/feature-registry.yaml \
  skills/reverse-engineer-rpi/fixtures/cc-sdd-v2.1.0/feature-registry.yaml

# 3. Update the pinned clone metadata
cp .tmp/cc-sdd-refresh/clone-metadata.json \
  skills/reverse-engineer-rpi/fixtures/cc-sdd-v2.1.0/clone-metadata.json

# 4. Commit the updated fixtures
git add skills/reverse-engineer-rpi/fixtures/cc-sdd-v2.1.0/
git commit -m "fix(reverse-engineer-rpi): update cc-sdd golden fixtures to <new-tag-or-sha>"

Fixture files that must be committed for the test to pass:

• skills/reverse-engineer-rpi/fixtures/cc-sdd-v2.1.0/clone-metadata.json
• skills/reverse-engineer-rpi/fixtures/cc-sdd-v2.1.0/feature-registry.yaml
• skills/reverse-engineer-rpi/fixtures/cc-sdd-v2.1.0/cli-surface-contracts.txt
• skills/reverse-engineer-rpi/fixtures/cc-sdd-v2.1.0/docs-features.txt

Script-Driven Workflow

Run:

python3 skills/reverse-engineer-rpi/scripts/reverse_engineer_rpi.py <product_name> --authorized [flags...]

This generates the required outputs under output_dir/ and (when applicable) .agents/council/ and .agents/learnings/.

Outputs (MUST be generated)

Core outputs under output_dir/:

1. feature-inventory.md
2. feature-registry.yaml
3. validate-feature-registry.py