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/bostonorange/claude-code-framework/ccf-debugging-playbooknpx skills add BostonOrange/claude-code-framework --skill ccf-debugging-playbookgit clone --depth 1 https://github.com/BostonOrange/claude-code-frameworkWhat 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 | $0.00153 | $0.03714 |
| Opus 5 | $0.00077 | $0.01857 |
| Sonnet 5 | $0.00031 | $0.00743 |
| Haiku 4.5 | $0.00015 | $0.00371 |
Grade B, and why
ccf-debugging-playbook scanned grade B with 1 finding 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 2d 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.
Reads agent configuration directoriesmediumAgent snooping
.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.
grep -rn "{{[A-Z_]*}}" .claude/ CLAUDE.md How it starts
The opening of the file, as written. The whole thing — 139 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CCF Debugging Playbook
Symptom → root cause triage for this repo's real failure modes. Every command below was verified against the repo as of 2026-07-11. Line numbers drift — re-verify via Provenance before trusting them.
Usage
Load this when something is broken and you don't yet know why: a red test in tests/, CI/local disagreement, a broken install in a target project, or a misbehaving hook. Once diagnosed, switch to the sibling that owns the fix (named per row below). NOT for "am I done?" gating (ccf-testing-and-qa), doc-count repair (ccf-doc-sync-campaign), or setup-script feature work (ccf-parity-playbook).
Process
- Reproduce narrowly. Run the single failing check:
bash tests/check-<name>.sh. Every check prints per-item PASS/FAIL detail — read it. - Classify via the triage table. If two causes fit, run the discriminating experiment in the topic sections.
- Fix at the root, together. Never hand-edit a count, allowlist line, or doc entry in isolation — change it together with the files it describes.
- Rerun the full gate:
bash tests/run-all.sh(8 bash checks; the PowerShell smoke is NOT in it — see "Local green, CI red").
Triage Table
| Symptom | First check | Likely cause | Fix / owning skill |
|---|---|---|---|
check-dogfood-drift: "Unexpected .claude/template drift" |
Did you (or a parallel session) add/edit anything under .claude/agents, hooks, skills, settings.local.json, or statusline? |
Real drift with no line in config/dogfood-drift-allowlist.txt (format = the exact diff output line, e.g. Only in .claude/skills: <name>) |
Sync the file with templates//skills/, or add the exact printed line with justification — ccf-dogfood-and-drift |
check-dogfood-drift: "Stale allowlist entries" |
Was a .claude/ file removed or synced back to its template? |
Allowlist line describes drift that no longer exists | Delete the stale line — ccf-dogfood-and-drift |
check-consistency FAILED |
The FAIL line names actual vs documented count and file (setup.sh or README.md) |
Something was added/removed without updating the summary phrases (N workflow skills, N AI agents, …), or a count was hand-edited |
Update files + counts together across all surfaces — ccf-doc-sync-campaign |
check-consistency shows documented="?" |
Was a summary line reworded? | The check greps exact phrases; rewording breaks extraction even if the number is right | Restore the phrase or extend the grep in tests/check-consistency.sh |
check-agent-registry FAILED |
Which of its 3 sub-checks failed | config/agents.json out of lockstep with templates/agents/*.md frontmatter, or agent name absent from README.md / templates/CLAUDE.md.template / docs/teams.md / docs/agents-commands-rules.md |
Update registry + frontmatter + all four docs together — ccf-doc-sync-campaign |
check-placeholders FAILED |
The FAIL line says which script lacks the token | New {{TOKEN}} has no replacement in setup.sh, setup.ps1, or config/placeholders.json (JSON counts as present for both) |
Add to BOTH scripts or to placeholders.json — ccf-parity-playbook |
| Setup smoke FAILED | The captured tmp-dir output the test prints on failure | Real setup.sh regression OR a harness bug | Run setup manually in a throwaway dir (below) |
| Suite green locally, CI red (or vice versa) | command -v python3; git --version; git config --get commit.gpgsign |
python3 absence, git < 2.28, host signing config, or the ps1-only CI job | "Local green, CI red" below |
Leftover {{TOKEN}} in an installed project |
Which file family contains it | One of three replacement sites failed | "Tracing leftover placeholders" below |
| Hook misbehaving / never firing | Pipe JSON manually (below); check the event name in settings | Bad stdin parsing, missing jq/python3, or invalid event name | "Hook debugging" below |
| Setup "succeeded" but project half-configured | grep Traceback on captured setup output; validate config/*.json |
The eval "$(python3 …)" set -e blind spot |
"Silent half-configuration" below |
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.
- 2d ago First seen · 139 lines · 153 tokens per session scan B 40d051615be3
ccf-debugging-playbook is a skill published in the GitHub repository BostonOrange/claude-code-framework (2 stars, last pushed 1mo ago), licensed MIT. It adds 153 tokens to every session and 3,714 once invoked, about $0.0008 per session on Opus 5. A static security scan graded it B with 1 finding (reads agent configuration directories). 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.
brainstorming
You MUST use this before any creative work - creating features, building components, adding functionality, or modifying behavior. Explores user intent, requirements and design before implementation.
chat-pet-sprite-creation
Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.
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…
agent-host-chat-contributions
Build and review cross-cutting agent-host chat behavior through lifecycle contributions. Use when adding turn lifecycle side effects, prompt or context injection, restored-history transformation, protocol-action observation, or when reviewing changes that add code to AgentSideEffects or AgentService.
auto-perf-optimize
Run agent-driven VS Code performance or memory investigations. Use when asked to launch Code OSS, automate a VS Code scenario, run the Chat memory smoke runner, capture renderer heap snapshots, take workflow screenshots, compare run summaries, or drive a repeatable scenario before heap-snapshot analysis.