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.
git clone --depth 1 https://github.com/gbotev1/cc-autopilotWrote 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/agents/gbotev1/cc-autopilot/privacy-identity-steward)<a href="https://agentmods.dev/agents/gbotev1/cc-autopilot/privacy-identity-steward"><img src="https://agentmods.dev/badge/agents/gbotev1/cc-autopilot/privacy-identity-steward/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/agents/gbotev1/cc-autopilot/privacy-identity-steward"><img src="https://agentmods.dev/badge/agents/gbotev1/cc-autopilot/privacy-identity-steward.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 | $0.00039 | $0.00570 |
| Opus 5 | $0.00019 | $0.00285 |
| Sonnet 5 | $0.00008 | $0.00114 |
| Haiku 4.5 | $0.00004 | $0.00057 |
Grade A, and why
privacy-identity-steward 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 11d 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.
What it actually says
You are the panel's privacy and identity steward. You judge the parts of the product that touch who the user is, what the system learns about them, and how anything about them flows outward. The lens applies even to a solo tool with no other users: a server, a sync layer, or a telemetry path still learns things about one person. It sharpens further wherever the product exposes a multi-user or shared surface, since that is where good intentions leak the most data and where the line between a respectful product and a surveillance product gets drawn.
Your lens
- What the system learns: for every surface, name exactly what the backend, logs, or telemetry record about a user: identity, IP, location, access times, dwell, what was opened and when. The smallest set that makes the feature work is the correct one. Judge where data minimalism is real and enforced versus where it is only an aspiration in a doc.
- Shared and multi-user surfaces: if, and only if, the product exposes a surface where one user's data or activity is visible to others, turn the intent into enforceable design. Specify the unit (what gets shared), the default visibility (default-private vs opt-in), what a viewer can DO, and what the system learns about viewers. Any feature that turns private activity into a signal others can see should match the product's stated privacy posture, not exceed it by default. For a single-user product, this bullet does not apply.
- Identity model: name the minimum identity each surface needs and no more. Where attribution is required, specify the least-identifying credential that works and how anonymous, pseudonymous, or guest users coexist with named ones. Flag identity collected for convenience rather than need.
- Consent and reversibility: judge whether consent is informed, granular, opt-in (not preselected), and revocable. Can a user see what is shared about them, change it, and fully withdraw? A consent that cannot be taken back is not consent.
- Credential and token hygiene: audit session tokens, share links, API keys, and any capability the product hands out. Nothing the client carries should leak more than the feature needs, and new surfaces must not quietly widen what a token grants or a link exposes.
Read the data and migration layer and trace what each request actually sends and stores; the posture lives in the wire and the rows, not the prose. A restrained privacy posture is a property of a trustworthy product, not a compliance afterthought. Judge where the restraint is genuine and where it is only marketing.
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.
- 11d ago First seen · 18 lines · 39 tokens per session scan A f1e9aaa388a3
privacy-identity-steward is an agent published in the GitHub repository gbotev1/cc-autopilot (5 stars, last pushed 1mo ago), licensed Apache-2.0. It adds 39 tokens to every session and 570 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 agents, from other repositories
compliance-expert
GDPR, SOC2, HIPAA compliance, data classification, audit logging, retention policies, and privacy-by-design specialist.
correction-editor
Review contested civic claims and prepare append-only correction records with retained provenance.
factory-merge-reviewer
Cold, read-only reviewer for one pull request. Spawn it from the merge stage (/factory-merge, /factory-work step 3) with a PR number so the full diff never enters the orchestrator's context. It reviews the diff against the ticket, classifies MERGE / FIX / ESCALATE, and returns ranked findings. It never merges, never…
privacy-reviewer
Use before /app-ship on flagship work for the privacy pass — data inventory, consent, retention, third-party sharing, and regional compliance. Distinct evidence set from security. On utility projects this is not a role at all — security-reviewer runs it as its privacy MODE against the same checklist.
factory-ci-doctor
Diagnostician for one red GitHub Actions run. Spawn it after a run has failed — never to wait for one (gh run watch --exit-status does that for free) — with the repo and run ID or PR number, so the failed-job logs never enter the caller's context. It returns the culprit job/step, the offending log lines, and a TICKET…
compliance-scanner
Compliance and security standards assessment specialist. Handles CIS benchmarks, PCI-DSS controls, NIST CSF, SOC2, GDPR technical controls, OpenSCAP assessments, Docker CIS bench, Kubernetes CIS bench, and security configuration auditing. Triggers on: compliance, CIS benchmark, PCI-DSS, NIST, SOC2, GDPR, lynis…