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 hariharapanigrahy/layerkit --skill layerkit-privacy-reviewgit clone --depth 1 https://github.com/hariharapanigrahy/layerkitWrote 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/hariharapanigrahy/layerkit/layerkit-privacy-review)<a href="https://agentmods.dev/skills/hariharapanigrahy/layerkit/layerkit-privacy-review"><img src="https://agentmods.dev/badge/skills/hariharapanigrahy/layerkit/layerkit-privacy-review/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/hariharapanigrahy/layerkit/layerkit-privacy-review"><img src="https://agentmods.dev/badge/skills/hariharapanigrahy/layerkit/layerkit-privacy-review.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 | $0.00030 | $0.00462 |
| Opus 5 | $0.00015 | $0.00231 |
| Sonnet 5 | $0.00006 | $0.00092 |
| Haiku 4.5 | $0.00003 | $0.00046 |
Grade A, and why
layerkit-privacy-review 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 10d 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 — 43 lines — stays where its author put it; the contents beside it link to each section on GitHub.
layerkit-privacy-review
Privacy review is source-code review plus evidence capture. Layerkit does not provide a runtime privacy gate for the client package.
Protocol
- Inventory new or changed fields from the vendor contract and client mapper.
- Classify likely PII/regulated data from field meaning, customer code, and policy evidence.
- Inspect existing consent, hashing, redaction, allowlist, denylist, and region checks in the client package.
- Update existing privacy code/tests directly when the vendor change requires it.
- Move raw API keys, passwords, private keys, bearer tokens, and credentials to environment variables or the client's secrets manager. In public or shared repos, source string literals that contain secrets are release blockers. Do not paste them into Layerkit memory/proposals/tests.
- Leave a TODO only when customer policy or datalayer support is missing.
- Record a redacted memory note:
layerkit memory append --type privacy --title "privacy review <vendor>" --vendor <vendor> --body-file ./privacy-digest.md
- Run the client package tests that cover privacy behavior and
layerkit doctor.
Forbidden
- Inventing legal basis, consent meaning, or privacy classification.
- Pasting real PII into Layerkit memory, proposals, tests, or docs.
- Leaving API keys, passwords, or tokens as source string literals.
- Shipping public/shared code while secret-like literals remain in source.
- Adding a parallel privacy layer when the existing client privacy path can be changed.
- Treating a Layerkit proposal as production privacy enforcement.
- Self-approving a privacy-sensitive change in strict maker-checker mode.
Success Criteria
- Changed PII fields are covered by existing or updated client checks.
- Privacy tests were updated when behavior changed.
- Secret-like values are env/SecretRef/secrets-manager references, not literals.
- Digest names residual human questions without raw PII.
- Unsupported data/policy gaps are explicit TODOs in the integration path.
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.
- 10d ago First seen · 43 lines · 30 tokens per session scan A 4d632c48afaf
layerkit-privacy-review is a skill published in the GitHub repository hariharapanigrahy/layerkit (8 stars, last pushed 1mo ago), licensed MIT. It adds 30 tokens to every session and 462 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 2026-08-31.
Other skills, from other repositories
yuwen-publish-precheck
A Chinese-language review workflow for checking content before publishing it on Douyin, Xiaohongshu, or WeChat Channels.
civic-publish-gate
Run deterministic source, citation, freshness, privacy, meeting-certainty, accessibility, correction-state, and human-approval checks before civic publication.
telnyx-email-suppressions-curl
Manage email suppressions (blocks), import and export suppression lists, and manage unsubscribe groups. Use for deliverability compliance and bounce handling.
editorial-correction-review
Review a challenged civic claim and prepare a visible, append-only correction or withdrawal with downstream propagation evidence.
healthcare-providers-verify
Validates practitioner credentials and license status against the NPI registry. Cross-references specialties, credentials, and practice addresses against official records. Returns Verified / Partially Verified / Unverified / Flagged per practitioner with mismatch details and source URLs. Triggers: "verify these…
python-packaging-license-finder
Use this skill to deterministically find license information for Python packages by checking PyPI metadata first, then falling back to Git repository LICENSE files using shallow cloning.