Borrowing it
Nothing to install: this file belongs to paullukic/coograph. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/paullukic/coograph/main/.github/skills/coograph-retro/SKILL.mdgit clone --depth 1 https://github.com/paullukic/coographWrote 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/skills/paullukic/coograph/coograph-retro)<a href="https://agentmods.dev/skills/paullukic/coograph/coograph-retro"><img src="https://agentmods.dev/badge/skills/paullukic/coograph/coograph-retro/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/skills/paullukic/coograph/coograph-retro"><img src="https://agentmods.dev/badge/skills/paullukic/coograph/coograph-retro.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 | $0.00066 | $0.04688 |
| Opus 5.5 | $0.00026 | $0.01875 |
| Sonnet 5.5 | $0.00013 | $0.00938 |
| Haiku 4.5 | $0.00007 | $0.00469 |
Grade A, and why
coograph-retro 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.
How it starts
The opening of the file, as written. The whole thing — 240 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Run a retro: measure how the guardrails held up across past sessions, then propose concrete changes as an OpenSpec. You propose. You never apply. Nothing outside openspec/changes/<date>-retro-<n>/ is written.
Read .github/retro/README.md once if you have not this session. It defines every detector, the registry, and the thresholds.
Step 0: Where do we stand
Run:
python3 .github/retro/retro.py --status
A layout: line may come first. It reports the instruction layout and never changes the exit code. When it says STRUCTURAL, tell the user /coograph-docs-restructure is the fix, whatever the exit code, before stopping on any row below. A project with no captured signals still needs to hear it.
Branch on the exit code and the retro: line printed:
| exit | line | what to do |
|---|---|---|
| 2 | _coograph_signals.py not found, or the command itself cannot be found |
.github/retro/ is not installed in this project. Tell the user to run /coograph-init (it copies .github/retro/ for every tool) or, for a registered project, to git pull coograph so sync.py delivers it. Stop. Write nothing. |
| 4 | not enabled; N archived changes found, bootstrap available |
Go to Bootstrap below, then continue at Step 1. |
| 3 | not enabled |
Fewer archived changes than bootstrap_min_archives. Say Retro is not enabled here, that init enables it, and stop. Write nothing. |
| 3 | no signals captured |
Offer a backfill (see Bootstrap, part 2 only). If the user declines or it captures nothing, stop. |
| 3 | N sessions since last retro ... |
Evidence is thin. Say so with the numbers and ask whether to proceed anyway. Proceed only on a yes. |
| 0 | N sessions since last retro ... |
Continue at Step 1. |
Bootstrap
Part 1, registry. Only when rules.json is absent:
python3 .github/retro/retro.py --merge-seed
This copies .github/retro/rules.seed.json (always shipped next to retro.py) to rules.json. Confirm it printed retro: registry ... (added graph-first, ...).
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 Changed · +6 lines 2f0853d61b58
- 1mo ago Changed · +5 lines 9be9055e4042
- 1mo ago Changed · +26 lines c1239ae1ca9a
- 1mo ago First seen · 203 lines · 66 tokens per session scan A 91a3cbbc97a1
coograph-retro is a skill published in the GitHub repository paullukic/coograph (17 stars, last pushed 12d ago), licensed MIT. It adds 66 tokens to every session and 4,688 once invoked, about $0.0003 per session on Opus 5.5. 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 2026-09-20.
coograph-retro compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| coograph-retropaullukic/coograph | Skill | 17 | 12d ago | MIT | A | 66 | Claude Code |
| managing-mcp-serversJetBrains/thinkrail | Skill | 7.8k | yesterday | Apache-2.0 | A | 62 | Claude Code, Codex |
| todosJetBrains/thinkrail | Skill | 7.8k | yesterday | Apache-2.0 | A | 58 | Claude Code, Codex |
| writing-workflow-skillsJetBrains/thinkrail | Skill | 7.8k | yesterday | Apache-2.0 | A | 60 | Claude Code, Codex |
| ai-ml-developmenttravisjneuman/.claude | Skill | 107 | yesterday | MIT | A | 43 | Claude Code, Codex |
| performance-engineeringtravisjneuman/.claude | Skill | 107 | yesterday | MIT | A | 38 | Claude Code, Codex |
| case-interview-practicetravisjneuman/.claude | Skill | 107 | yesterday | MIT | A | 39 | Claude Code, Codex |