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-debug-memorygit 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-debug-memory)<a href="https://agentmods.dev/skills/potpie-ai/potpie/potpie-debug-memory"><img src="https://agentmods.dev/badge/skills/potpie-ai/potpie/potpie-debug-memory/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-debug-memory"><img src="https://agentmods.dev/badge/skills/potpie-ai/potpie/potpie-debug-memory.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 | $0.00041 | $0.01468 |
| Opus 5.5 | $0.00016 | $0.00587 |
| Sonnet 5.5 | $0.00008 | $0.00294 |
| Haiku 4.5 | $0.00004 | $0.00147 |
Grade A, and why
potpie-debug-memory 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 3d 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 — 136 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Potpie Debug Memory
Use this skill before digging into a failure so prior symptoms, known fixes, and failed attempts can guide the investigation.
Fast Path
Share one discovery pass across relevant skills; reuse current reads and hook
context for the same task, pot, and scope instead of repeating resolve. Run
resolve alongside scoped preferences for code work and an untyped
search-entities lookup when the task names a service whose key is unknown.
Use a known explicit pot selector; resolve ambiguous routing first and check
returned pot IDs before combining results. Local file discovery can run alongside
these reads. Use returned keys for later scoped reads rather than guessing.
Search by symptom, not just component name: exact error text, failing test,
command, environment, service, dependency, and synonyms. The shared resolve call
infers intent from words like why / broken / stale / failing / error and returns
prior bugs, the timeline and infra as triples (--intent debugging pins it):
potpie resolve "<symptom in the user's words, plus the exact error text>"
Runbooks and their recovery steps live in ingested documents and notes, which
prior_occurrences never returns. Choose the includes for the initial resolve
call rather than running both resolve examples. If it lacks document evidence,
read a matching section, then fetch returned chunk IDs in one batched call:
potpie resolve "<symptom>" --include prior_bugs,docs,timeline
potpie graph read --subgraph knowledge --view document_context --query "<symptom>" --limit 5
potpie resource get potpie://res/<doc>/<section>/0000 --with-neighbors
Go to the named view only when resolve is thin. Expand the query there with
synonyms and the exact error text; the view ranks its pool and returns up to
--limit rows however weak (--query-threshold does nothing here), so judge
each row by its score and text — an exact hit can score as low as 0.55:
potpie graph read --subgraph debugging --view prior_occurrences --query "<expanded symptom query>" --scope service:<service-name> --limit 12
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.
- 3d ago Changed · +15 lines c1bad3bc29ea
- 7d ago Changed · +41 lines b51fd1df6843
- 1mo ago First seen · 80 lines · 41 tokens per session scan A f3fdab1915bd
potpie-debug-memory is a skill published in the GitHub repository potpie-ai/potpie (5,743 stars, last pushed today), licensed Apache-2.0. It adds 41 tokens to every session and 1,468 once invoked, about $0.0002 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
compiler-11897630
A coverage skill that marks or reports how much of a set of instructions is covered by tests. Test coverage is a measure of which code paths tests exercise.
memory
Use memory-as-tool to add and search memories throughout the Codex session.
analyzing-all-conversations
Produce a coverage-checked report of the entire Memex conversation corpus when the user asks to analyze, organize, or summarize all Codex history. Use deterministic totals first and label unfinished backfill honestly.
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.
mcp-apps-builder
MANDATORY for ALL MCP server work - mcp-use framework best practices and patterns. READ THIS FIRST before any MCP server work, including: Creating new MCP servers Modifying existing MCP servers (adding/updating tools, resources, prompts, widgets) Debugging MCP server issues or errors Reviewing MCP server code for…