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.
npx agentmods add skills/klausfreiberufler/devflow-mcp/devflow-debuggingnpx skills add KlausFreiberufler/devflow-mcp --skill devflow-debugginggit clone --depth 1 https://github.com/KlausFreiberufler/devflow-mcpWhat 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 | $0.00061 | $0.01305 |
| Opus 5 | $0.00030 | $0.00652 |
| Sonnet 5 | $0.00012 | $0.00261 |
| Haiku 4.5 | $0.00006 | $0.00130 |
Grade A, and why
devflow-debugging 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 yesterday.
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 — 125 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Skill: devflow-debugging
Iron-Law port of Superpowers
systematic-debugging. The four phases are non-negotiable; skipping ahead leaves untested assumptions in the code base.
Sister skill (DF-316): run
devflow-error-investigatorFIRST to pull wiki signals relevant to the error (runbooks, affected ADRs, recent flows, similar past errors). Then come back here for the systematic 4-phase walk. Together they enforce: never debug from scratch what the wiki has already seen.
When to use
- Test fails unexpectedly (not the planned RED of
devflow-tdd) - Behavior on staging/prod doesn't match the spec or the test-suite
- Reviewer flags a regression you can't immediately explain
Skip when: the failure is the expected RED of a TDD cycle — that belongs to devflow-tdd.
Process
1 · Investigation
Capture the bug surface area before forming any opinion:
- What was observed? Exact error message, stack-trace, screenshot, or test output.
- What was expected? Acceptance-criterion or contract that says it should be different.
- Reproduction recipe? Commands or steps that trigger the bug deterministically.
- First time? Or regression? Check git-blame on the suspected file; check the changelog.
If you cannot reproduce on demand: stop. Make reproduction work first — without it the next phases are guesswork.
2 · Pattern
Look for shape-matches in what you've seen before:
- Have I (or another flow) seen this stack-trace? Search the project's flows + agent-logs.
- Does the error message match a known runbook? → Hook 4 (next section).
- Does the failure mode (race, off-by-one, encoding, auth) suggest a class of bugs?
Hook 4 · Runbook Match (BEFORE Hypothesis)
GET /api/wiki/search?q=<error-message>&types=doc_page&document_type=runbook
Equivalent MCP: wiki_search({ q: errorMessage, types: ['doc_page'], document_type: 'runbook' }).
If a runbook matches:
- Read it through. Symptoms, diagnosis, fix-steps.
- If the fix-steps clearly apply → skip to Phase 4 (Fix) with a wiki-link
[[runbook:<slug>]]in your commit message. Don't re-derive what the runbook already documents. - If the runbook is related but not exact → cite it in the discipline-token evidence as a partial match and continue to Phase 3.
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.
- yesterday First seen · 125 lines · 61 tokens per session scan A 65cb2f1fc122
devflow-debugging is a skill published in the GitHub repository KlausFreiberufler/devflow-mcp (1 stars, last pushed 3d ago), licensed MIT. It adds 61 tokens to every session and 1,305 once invoked, about $0.0003 per session on Opus 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 2026-08-31.
Other skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
babysit-pr
Babysit a GitHub pull request after creation by continuously polling review comments, CI checks/workflow runs, and mergeability state until the PR is merged/closed or user help is required. Diagnose failures, retry likely flaky failures up to 3 times, auto-fix/push branch-related issues when appropriate, and keep…
imagegen
Generate or edit raster images when the task benefits from AI-created bitmap visuals such as photos, illustrations, textures, sprites, mockups, or transparent-background cutouts. Use when Codex should create a brand-new image, transform an existing image, or derive visual variants from references, and the output…
cpu-profile-analysis
Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…
next-cache-components-optimizer
Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…