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/tradebaas/groundwork/debugnpx skills add Tradebaas/Groundwork --skill debuggit clone --depth 1 https://github.com/Tradebaas/GroundworkWhat 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 | $0.00080 | $0.00696 |
| Opus 5 | $0.00040 | $0.00348 |
| Sonnet 5 | $0.00016 | $0.00139 |
| Haiku 4.5 | $0.00008 | $0.00070 |
Grade A, and why
debug 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 — 50 lines — stays where its author put it; the contents beside it link to each section on GitHub.
debug: from symptom to proven root cause
A fix you never saw fail is a guess (verify). This loop turns an unknown failure into a
proven fix at the lowest cost. Production incident? Stabilize first per maintain (rollback
beats diagnosis under fire), then run this loop calmly.
The loop
- Read the actual error. The message, the stack, the failing line. Capture expected vs actual behavior verbatim before touching anything: the error usually says more than the first theory about it.
- Reproduce before fixing. A failing automated test at the nearest seam is the goal; a minimal script is the fallback. Shrink the repro until removing anything makes the bug disappear. What you cannot reproduce you cannot fix, only disturb.
- Check what changed. Most bugs live in the newest diff: recent commits, dependency
bumps, config or environment changes.
git bisectbeats rereading the whole system. - Hypothesize, test one change at a time. Rank hypotheses by likelihood and cost of testing; each experiment changes exactly one thing and names its predicted outcome before running. Two simultaneous changes make the result unreadable. Expect horses, not zebras: the common cause outranks the exotic one until evidence says otherwise.
- Fix the root cause, once. The fix goes where the cause lives, not where the symptom showed (decision ladder: understand, then fix the cause once). A symptom patch that leaves the cause in place is not progress, and per debug hygiene a change that demonstrably fixed nothing gets reverted, not left in.
- Prove and clean. The repro test passes and stays as the regression test; every debug shim (prints, sleeps, forced branches, commented-out calls) is removed; the full gate chain runs green.
On a platform stack
When the stack file's header declares a hosted platform, steps 2 and 3 keep their intent and change their instrument; the rest of the loop is unchanged. Reproduce in a real dev environment, against the platform's own record of the failure - run history, execution logs, the audit trail - and with its test tooling where the platform has any. For what changed, the platform keeps its own history: solution layers and version history, the audit log of who edited what, and the vendor's release notes for the change nobody in the project made. That is the bisect equivalent, and it reaches one thing bisect cannot - an edit made directly in the platform editor that never became a commit.
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 · 50 lines · 0 tokens per session scan A b688bc746995
debug is a skill published in the GitHub repository Tradebaas/Groundwork (2 stars, last pushed 6d ago), licensed MIT. It adds 80 tokens to every session and 696 once invoked, about $0.0004 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
mission-driver
Create roadmaps and mission configs, then drive the mission-driver AI dev-loop engine. mission-driver lives at tools/mission-driver/ — it reads missions/ .json and loops CHECK → REVIEWPLANS → EXECPLANS → DRAFTPLANS → DEEPAUDIT until the roadmap is done or the audit budget is exhausted. Use this skill when the user…
keel-plan
Create or revise PLAN.md — propose the phase DAG (phases · gates · dependencies) from the user's goal, get approval, write the table and regenerate the colored Mermaid diagram from it. Statuses flip at rituals (/keel-handover, /keel-phase-review); post-completion fixes land in the Fix log.
keel-stats
Render the ritual telemetry visually — turns .claude/ritual-log into reports/ritual-stats.md with PLAN.md-style colored Mermaid interval boxes (session/compact boundaries) + a counts table. Answers "which skills/commands/hooks ran, how often, in which interval".
init
Configure an uninitialized project from an approved charter: stack, scripts, identity, optional Git, and verification.
sdd-serve
Serve the SDD Builder's AI request queue: claim requests with sddnextrequest, draft the proposal, answer with sddrespondrequest. Never writes spec files — the user accepts each proposal in the builder. Use when the user asks to attend, serve or listen to the SDD board queue. / Atiende la cola de peticiones del SDD…
review-creative
Audit customer-facing work for UX, accessibility, brand fit, credibility, and generic AI output. Use for design, marketing, document, presentation, or visual review.