Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
git clone --depth 1 https://github.com/rtazima/claude-proj-blueprintWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/commands/rtazima/claude-proj-blueprint/discover)<a href="https://agentmods.dev/commands/rtazima/claude-proj-blueprint/discover"><img src="https://agentmods.dev/badge/commands/rtazima/claude-proj-blueprint/discover/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/commands/rtazima/claude-proj-blueprint/discover"><img src="https://agentmods.dev/badge/commands/rtazima/claude-proj-blueprint/discover.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00000 | $0.00662 |
| Opus 5.5 | $0.00000 | $0.00265 |
| Sonnet 5.5 | $0.00000 | $0.00132 |
| Haiku 4.5 | $0.00000 | $0.00066 |
Grade A, and why
discover scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 1mo ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
What it actually says
Legacy code archaeology and context bootstrap. Reads code, git history, Jira (when configured), and remote PRs/issues to reconstruct WHY a module exists. Produces discovery notes in memory/, ADR candidates and PRD candidates in docs/ for human review, then re-indexes memory.
Arguments: $ARGUMENTS (one or more module paths or file paths, space-separated)
Workflow:
- Activate the
context-legacyskill - For each target in $ARGUMENTS, run all phases in order:
- Phase 0 — Memory query: check if already discovered, avoid redo
- Phase 1 — Code archaeology: static analysis, hot/cold files, public interface, surprising patterns
- Phase 2 — Git history: churn, authors, regression commits, decision messages, git blame (targeted on hot files), dead PRs (closed without merge)
- Phase 3 — Jira enrichment: ticket history, bug patterns, decision comments, epic chain (skipped if
JIRA_*not set in.env) - Phase 4 — Cross-reference: correlate all signals, produce synthesis table
- Phase 5 — Output: discovery notes, main context doc, ADR candidates, PRD candidates, marker patch, re-index
- Never modify source files — read-only on code
Tunable flags (defaults in parentheses, override via env var):
LEGACY_CONTEXT_EARLY_EXIT(false): if true, Phase 2 dead PR scan may run in lean mode on repos with no PR review cultureLEGACY_CONTEXT_TRUNK_THRESHOLD(30): PR coverage % below which the repo is classified as trunk-based
Defaults bias toward completeness. Override for speed when you have validated the repo profile.
Output paths:
- Survey: memory/discoveries/-survey.md
- Per-file notes: memory/discoveries//.md
- Dead PR notes: memory/discoveries/-dead-prs.md
- Jira notes: memory/discoveries/-jira.md
- Main context doc: docs/architecture/legacy-context--.md
- ADR candidates: docs/architecture/_candidates/adr-candidate-.md
- PRD candidates: docs/product/_candidates/prd-candidate-.md
- Marker patch: memory/discoveries//markers.patch
After all targets are processed, run: memory/.venv/bin/python -m memory.index --incremental
Supported remote providers (Phase 2 dead PR scan):
- GitHub: detected by
github.comin remote URL. Requiresgh auth logincompleted. - Azure DevOps: detected by
dev.azure.comor.visualstudio.com. RequiresAZURE_DEVOPS_EXT_PATin shell. - Other providers (GitLab, Bitbucket, self-hosted): dead PR scan skipped with warning.
Use this before /implement or /refactor on modules where memory and docs are empty. Designed for codebases older than 5 years with no convention discipline.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 1mo ago First seen · 41 lines · 0 tokens per session scan A 12c83127d504
discover is a command published in the GitHub repository rtazima/claude-proj-blueprint (22 stars, last pushed 5mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 662 tokens. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other commands, from other repositories
sruja-pre-merge-review
Structured review workflow before merging changes. Produces actionable feedback with architecture grounding.
review
Architecture-aware code review with Pharaoh.
restore
/restore - Restore Git & Session Context.
brief
Brief a PR, MR or the current branch with Punchcard — the reviewer's orientation before the review, never a review.
pr
Review a GitHub PR or GitLab MR with Punchcard and post the review into it.
spec-release
Generate release notes, changelog, and deployment checklist.