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 skills add ww-w-ai/bkit-claude-code --skill pdca-watchgit clone --depth 1 https://github.com/ww-w-ai/bkit-claude-codeWrote 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/ww-w-ai/bkit-claude-code/pdca-watch)<a href="https://agentmods.dev/skills/ww-w-ai/bkit-claude-code/pdca-watch"><img src="https://agentmods.dev/badge/skills/ww-w-ai/bkit-claude-code/pdca-watch.svg" alt="Measured on agentmods" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 2 findings, up to medium
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- medium analysis-evasion · line 1 Suspicious Unicode normalization or mixed-script contentFix: Review the flagged content for security risks. Ensure no credentials, secrets, or sensitive data are exposed.
- medium Excessive Agency · line 113 Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
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 | $0.00048 | $0.00991 |
| Opus 5 | $0.00024 | $0.00495 |
| Sonnet 5 | $0.00010 | $0.00198 |
| Haiku 4.5 | $0.00005 | $0.00099 |
Grade A, and why
pdca-watch 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 4d 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 — 116 lines — stays where its author put it; the contents beside it link to each section on GitHub.
PDCA Watch — Live Tick Dashboard
v2.1.11 Sprint β FR-β4. Wraps
lib/dashboard/watch.jsover CC's/loop(v2.1.71+). Read-only — only consumes state files; never writes. Falls back to a single render when/loopis unsupported (E-β4-01).
Arguments
| Argument | Description | Example |
|---|---|---|
| (none) | Auto-resolve active feature from pdca-status.primaryFeature |
/pdca-watch |
<feature> |
Watch a specific feature by name | /pdca-watch bkit-v2111-integrated-enhancement |
Behavior
Resolution
- If
<feature>is provided, use it verbatim. - Else read
.bkit/state/pdca-status.json→primaryFeature, thenactiveFeatures[0]as fallback. - If still no feature → render
"No active PDCA feature. Start one with /pdca pm <feature>"and exit.
Tick (every 30s via /loop)
For each tick, lib/dashboard/watch.js#renderTick produces a panel with:
Watch <feature> — tick N (HH:MM:SS)
─────────────────────────────────────────────────────────────
Phase: <phase> Match: <%> Iter: <n / 5>
Tokens: <in> in / <out> out · samples <N>
Est cost: $<USD>
─────────────────────────────────────────────────────────────
Data sources:
- Phase / matchRate / iterationCount:
pdca-status.features[<feature>] - Token totals: last 50 lines of
token-ledger.ndjson - Cost: Sonnet-class flat estimate ($3/Mtok in, $15/Mtok out)
Termination
- Ctrl-C from user
- Feature
phase === 'completed'orphase === 'archived' - After max 60 ticks (≈30 min) — caller restarts if more time needed
CC /loop Support
| CC Version | Behavior |
|---|---|
| ≥ 2.1.71 | Native /loop 30s ticker |
| < 2.1.71 | E-β4-01: single render + warning, no loop |
| Unknown | Same as < 2.1.71 (fail-safe) |
The check is performed via watch.checkLoopSupport({ ccVersion }) —
caller passes the resolved CC version from lib/infra/cc-version-checker.
Security
- Pure read on
.bkit/state/pdca-status.jsonand.bkit/runtime/token-ledger.ndjson. - Tail bounded to
MAX_TAIL_LINES = 200regardless of caller input — defends against pathological NDJSON growth. - No subprocess spawning, no network.
- PII redaction on the ledger is owned by the writer (Token Accountant); watch only reads what's already redacted.
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.
- 4d ago First seen · 116 lines · 48 tokens per session scan A d15384888653
pdca-watch is a skill published in the GitHub repository ww-w-ai/bkit-claude-code (595 stars, last pushed yesterday), licensed Apache-2.0. It adds 48 tokens to every session and 991 once invoked, about $0.0002 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 2026-09-03.
Other skills, from other repositories
magazine-web-ppt
For marketing and gtm work: bind launches, campaigns, events, and brand plans to growth and pipeline outcomes. Built around the core query "annual-marketing-plan", with GTM strategy lead judgment, buyer-ready proof, and this outcome: approve launch plan, campaign budget, or GTM motion.
html-ppt-zhangzara-broadside
OpenDesign's product-launch announcement and press narrative — the headline, the proof points, and the call to action. Built as a decision-grade marketing & GTM deck for press, community, prospects.
html-ppt-zhangzara-coral
OpenDesign's community-growth campaign across GitHub, Discord, and X: the loops, the content calendar, and the pipeline math. Built as a decision-grade marketing & GTM deck for growth team, community lead.
drug-design
End-to-end drug discovery pipeline orchestration. Deterministic Python script that auto-chains structure prediction, pocket detection, de novo design, docking, scoring, and ADMET filtering into reproducible workflows.
hunt-cicd
Hunt CI/CD pipeline vulnerabilities — GitHub Actions workflow injection (pullrequesttarget Pwnrequest + ${{ }}-into-shell), self-hosted runner poisoning, OIDC trust-policy abuse, Jenkins script-console RCE and CVE-2024-23897 file read, GitLab CI runner-token registration, Terraform state file leakage, artifact/log…
CI/CD Pipeline Advanced
Expert-level CI/CD pipeline skill for test automation. Covers GitHub Actions, Jenkins, GitLab CI, Azure DevOps, parallel execution, matrix strategies, caching, artifact management, and deployment gates.