Potpie is a tool that turns a software repository and its development history into a context graph for AI coding agents. Agents use that graph to answer codebase questions, plan changes, debug problems, and write code with project-specific information. The catalogue includes skills, hooks, commands, and a plugin for integrating Potpie into coding-agent workflows.
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 potpie-ai/potpie --skill potpie-graphgit clone --depth 1 https://github.com/potpie-ai/potpieWrote 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/potpie-ai/potpie/potpie-graph)<a href="https://agentmods.dev/skills/potpie-ai/potpie/potpie-graph"><img src="https://agentmods.dev/badge/skills/potpie-ai/potpie/potpie-graph/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/potpie-ai/potpie/potpie-graph"><img src="https://agentmods.dev/badge/skills/potpie-ai/potpie/potpie-graph.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00107 | $0.06909 |
| Opus 5.5 | $0.00043 | $0.02764 |
| Sonnet 5.5 | $0.00021 | $0.01382 |
| Haiku 4.5 | $0.00011 | $0.00691 |
Grade A, and why
potpie-graph 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 — 496 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Potpie Graph Workbench
Supported single-record types:
preference|policy|bug_pattern|fix|verification|decision|doc_reference|workflow|runbook_note|incident_summary|investigation|diagnostic_signal|service_note|feature_note|integration_note
The graph is project memory: preferences, prior bugs and their fixes, infra topology, decisions, a timeline of changes, and ingested documents. You are the intelligence that reads it before acting and writes durable learnings after. Potpie validates, lowers, commits, audits, and ranks. It does not scan a repository or infer rich facts from prose for you.
Text output for reads; --json for propose, commit, resource import,
and anything you parse. graph describe --examples prints its examples only
with --json, and they are read commands: the write payload shape is
graph mutation-template.
When a command's JSON is saved to a file, parse it with a JSON decoder (for
example, json.load) and print only the fields needed for the task. Do not
regex-match or reprint an entire minified JSON object to inspect one field.
Protocol contracts and decoder questions
Protocol memory is an opt-in extension, off unless the user turned it on with
potpie config set graph.protocols on (and restarted a running daemon). For
telegrams, message layouts, protocol revisions, field values or decoder changes,
check that the catalog advertises protocols.message_context; when it does, load
the shared protocol reference. Discovery is
potpie --json resolve '<question>' --include protocols --pot <pot>, followed by
the returned message reads. This replaces the default broad resolve in section 2
for these questions: ordinary includes omit protocols. When the catalog does not
advertise the view, the extension is off; say so rather than guessing a layout.
Apply this route when a generic status/value question turns out to concern a protocol in a fetched source, too. A document hit alone does not select a message. Before assigning an unscoped value meaning, show the matching message/revision alternatives and distinguish integer from string values. Reuse a scope already selected in the conversation, stating it in the answer. The reference covers sourced ingestion, compact follow-ups, corrections and coverage.
What ships with it
2 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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 Changed · +82 lines · +13 tokens per session bdc903c7260c
- 8d ago Changed · +178 lines 51e8fb330081
- 1mo ago First seen · 236 lines · 94 tokens per session scan A c5015c129adb
potpie-graph is a skill published in the GitHub repository potpie-ai/potpie (5,746 stars, last pushed 2d ago), licensed Apache-2.0. It adds 107 tokens to every session and 6,909 once invoked, about $0.0004 per session on Opus 5.5. 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.
Other skills, from other repositories
memory
Use memory-as-tool to add and search memories throughout the Codex session.
github
Interact with GitHub using the gh CLI. Use gh issue, gh pr, gh run, and gh api for issues, PRs, CI runs, and advanced queries.
langchain_patterns
Implement Retrieval-Augmented Generation (RAG) systems with LangChain4j. Build document ingestion pipelines, embedding stores, vector search strategies, and knowledge-enhanced AI applications. Use when creating question-answering systems over document collections or AI assistants with external knowledge bases.
langchain_patterns
Implement Retrieval-Augmented Generation (RAG) systems with LangChain4j. Build document ingestion pipelines, embedding stores, vector search strategies, and knowledge-enhanced AI applications. Use when creating question-answering systems over document collections or AI assistants with external knowledge bases.
project-graveyard
Scans the developer's machine for dead side projects, autopsies each one from its git history (died at the payments wall, killed by a newer project, finished but never shipped), surfaces their personal death patterns, and picks the corpse most worth resurrecting — then helps ship it. Use when the user mentions…
commit-archaeologist
Reconstructs why code exists from local git history, including the introducing commit, later changes, current authors, repeated companion files, and likely intent. Use when the user asks "why does this code exist", "who wrote this function and why", or to "explain the history of this function" before a rewrite…