What is review-leaks?

A security review guide for finding secrets and sensitive information in Agentgram code and configuration before a push. It covers credentials in YAML, environment files, authentication settings, databases, and deployment files.

Use it to review code and configuration for hardcoded secrets, incorrectly exposed environment variables, committed .env files, and other data leaks.

It helps catch passwords, tokens, connection details, and other private values that could be exposed through source control or deployment configuration.

How dfradehubs/agentgram configures Claude Code

Reuse

Borrowing it

Nothing to install: this file belongs to dfradehubs/agentgram. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file

curl -O https://raw.githubusercontent.com/dfradehubs/agentgram/main/.claude/skills/review-leaks/SKILL.md

Clone the repo

git clone --depth 1 https://github.com/dfradehubs/agentgram

Made for: Claude Code.

Wrote 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.

README.md

[![agentmods](https://agentmods.dev/badge/skills/dfradehubs/agentgram/review-leaks/github.svg)](https://agentmods.dev/skills/dfradehubs/agentgram/review-leaks)

Your own site

<a href="https://agentmods.dev/skills/dfradehubs/agentgram/review-leaks"><img src="https://agentmods.dev/badge/skills/dfradehubs/agentgram/review-leaks/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.

Your own site · 80×15

<a href="https://agentmods.dev/skills/dfradehubs/agentgram/review-leaks"><img src="https://agentmods.dev/badge/skills/dfradehubs/agentgram/review-leaks.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 19 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 767 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.

Origin original No closer match found in the catalogue.

Token cost

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	$0.00019	$0.00767
Opus 5	$0.00010	$0.00383
Sonnet 5	$0.00004	$0.00153
Haiku 4.5	$0.00002	$0.00077

Measured 10d ago against content hash a40caba61d89, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-09, from the pricing page.

Security

Grade A, and why

review-leaks 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.

.claude/skills/review-leaks/SKILL.md · 86 lines

How it starts

The opening of the file, as written. The whole thing — 86 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Act as a Security Engineer specialized in secret detection and data leak prevention for the Agentgram platform.

Critically review the code provided as if you were the last line of defense before pushing. Be paranoid, thorough, and explicit.

Project context

Config system: YAML files in api/configs/ use ${ENV:VAR} syntax for secrets. Any literal secret in YAML is a leak.
Auth: Keycloak OIDC — client IDs, client secrets, JWKS URLs, issuer URLs.
Database: PostgreSQL (POSTGRES_* env vars), Redis (REDIS_* env vars).
Registry: ghcr.io/dfradehubs/ (GitHub Container Registry).
Web: Next.js with NEXT_PUBLIC_* env vars (publicly exposed) vs server-only env vars.
K8s: Deployed to Kubernetes.

Evaluate

Hardcoded secrets

API keys, tokens, passwords, passphrases
Keycloak client secrets, OIDC credentials
JWT signing keys, JWKS material
PostgreSQL/Redis connection strings with credentials
GitHub OAuth tokens (used for agent token forwarding)
Values in YAML that should use ${ENV:VAR} but don't

Configuration files

.env files or .env.* variants committed
configs/*.yaml with literal credentials (even commented)
docker/docker-compose.yml exposing real secrets
web/.env.local or web/.env.production committed
NEXT_PUBLIC_* vars leaking server-side secrets to the browser

Internal infrastructure exposure

Keycloak issuer URLs pointing to internal/staging instances
PostgreSQL/Redis hostnames or connection strings
GCP project IDs, Artifact Registry paths
Internal service names, cluster details
K8s namespace or service discovery info

Personally Identifiable Information (PII)

Real emails, Google Workspace group names
Test data with real user information
Log output containing JWT claims or user data
Hardcoded user IDs or Keycloak subject IDs

Debug and development artifacts

auth.enabled: false left in production configs
LOG_LEVEL: debug in non-dev configs
POSTGRES_SSLMODE: disable in non-dev configs
Verbose logging exposing JWT contents or agent responses
TODO/FIXME comments with sensitive context

Read the full file on GitHub · 86 lines

Changes

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 · 86 lines · 19 tokens per session scan A a40caba61d89

Subscribe to this mod's changes

review-leaks is a skill published in the GitHub repository dfradehubs/agentgram (20 stars, last pushed 19d ago), licensed MIT. It adds 19 tokens to every session and 767 once invoked, about $0.0001 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 2026-08-30.

Security a2a ag-ui ai-agents mcp mcp-gateway

documentdb-collection-admin

Database, collection, and user administration on a DocumentDB (MongoDB-compatible, PostgreSQL-backed) server via the documentdb-mcp MCP server — list/create/drop/rename databases and collections, and manage database users and their roles. Use when the agent must provision or tear down namespaces, enumerate the…

Knuckles-Team/documentdb-mcp · 106 tokens

documentdb-document-crud

Document create/read/update/delete operations on a DocumentDB (MongoDB-compatible, PostgreSQL-backed) collection via the documentdb-mcp MCP server — insert one/many, find with filter/sort/skip/limit, update/replace one/many, delete, count, and the atomic find-and-modify variants. Use when the agent must read or mutate…

Knuckles-Team/documentdb-mcp · 126 tokens

jellyfin-kg-ingestion

Natively ingest a Jellyfin library into the epistemic-graph knowledge graph via the jellyfin-mcp MCP server — push library items as typed :MediaAsset/:Book nodes with :hasGenre/:performedBy/:authoredBy links and item overviews as searchable :Document nodes (jellyfiningestlibrary), and item posters as content-addressed…

Knuckles-Team/jellyfin-mcp · 158 tokens

jellyfin-library-catalog

Browse and search a Jellyfin media library via the jellyfin-mcp MCP server — list library items (movies, series, episodes, audio, books), fetch one item by id, walk collections and user views, and run search hints with the domain-typed jellyfinlibrary tool. Use when the agent must find or enumerate catalog items…

Knuckles-Team/jellyfin-mcp · 154 tokens

jellyfin-media-playback

Stream and control Jellyfin playback and browse music via the jellyfin-mcp MCP server — list artists, albums, and music genres, resolve audio/video streams, manage playlists, and drive playstate (play, mark played, report progress) with the domain-typed jellyfinmedia tool. Use when the agent must start or control…

Knuckles-Team/jellyfin-mcp · 142 tokens

documentdb-aggregation-analytics

Read-only analytics on a DocumentDB (MongoDB-compatible, PostgreSQL-backed) collection via the documentdb-mcp MCP server — run aggregation pipelines ($match/$group/$sort/…) and fetch distinct field values. Use when the agent must summarize, group, or profile documents rather than fetch individual rows. Do NOT use for…

Knuckles-Team/documentdb-mcp · 102 tokens

agentgram: Skill for Claude Code