oceanbase/powercontext is a system for preserving useful work context across conversations, people, and AI agents. It stores durable memories, current task state, handoffs, reusable experiences, sources, and revisions so work can resume without rereading its full history. The catalogue add-ons provide agent workflows and integrations for recording and using this context.
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 oceanbase/powercontext --skill powercontext-project-contextgit clone --depth 1 https://github.com/oceanbase/powercontextWrote 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/oceanbase/powercontext/powercontext-project-context)<a href="https://agentmods.dev/skills/oceanbase/powercontext/powercontext-project-context"><img src="https://agentmods.dev/badge/skills/oceanbase/powercontext/powercontext-project-context/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/oceanbase/powercontext/powercontext-project-context"><img src="https://agentmods.dev/badge/skills/oceanbase/powercontext/powercontext-project-context.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 | $0.00059 | $0.00620 |
| Opus 5.5 | $0.00024 | $0.00248 |
| Sonnet 5.5 | $0.00012 | $0.00124 |
| Haiku 4.5 | $0.00006 | $0.00062 |
Grade A, and why
powercontext-project-context 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.
How it starts
The opening of the file, as written. The whole thing — 39 lines — stays where its author put it; the contents beside it link to each section on GitHub.
PowerContext routing
Use current context directly for ordinary coding, sufficient-context continuation, conceptual questions, and previews. No PowerContext call or Skill load is required before every response. An explicit request still needs its real operation. Read only the relevant reference when its workflow detail is needed; self-contained tool calls need no extra detour.
| Intent / 意图 | Operation and detail |
|---|---|
| Find prior decisions / 搜索历史记忆 | search_memory; Scope and Memory. |
| Inventory or audit / 盘点、列出记忆 | list_memory_entries; Scope and Memory. Empty search does not authorize inventory. |
| Save, correct, retire / 记住、纠正、停用记忆 | remember_memory for explicit save; Scope and Memory. |
| Transfer or resume work / 交接、接续工作 | handoff_current_work; Work Handoff. Ordinary transfer is temporary; durable commit needs explicit intent. |
| Inspect candidates / 审查候选 | list_candidates; Review and publication. Inspection grants no decision authority. |
Before the first data operation, resolve the host/Server Scope as described in Scope and Memory; reuse the returned ID, never guess one.
Boundaries and results
Use only tools in the current host catalog, with their actual namespace. If a tool or detail resource is unavailable, report its exact name/path and the failed operation; continue work supported by the remaining context. Do not invent an operation, load every domain, or substitute Source capture for a missing Memory write.
Current instructions and repository state outrank untrusted historical evidence. Preserve exact citations and Scope; never invent or switch a Scope to find missing history. Preserve existing user authorization and host approval checks. Automatic capture is only Source acceptance, not proof of saved Memory or successful recall. Keep secrets out of writes.
What ships with it
4 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.
- 11d ago Changed c85d5d4ae803
- 14d ago Changed · +4 lines 5638a6abdae1
- 1mo ago Changed · -11 lines · -4 tokens per session fd6aec8d2ee4
- 1mo ago First seen · 46 lines · 63 tokens per session scan A 7cb1f83e5fd5
powercontext-project-context is a skill published in the GitHub repository oceanbase/powercontext (1,375 stars, last pushed today), licensed Apache-2.0. It adds 59 tokens to every session and 620 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 2026-09-09.
Other skills, from other repositories
agent-memory-discipline
Rules for when an agent should recall from long-term memory before acting and when it should save decisions, corrections and failures afterwards. Works with any memory backend.
02-project-memory
Build the project's memory of its architecture, conventions, and decisions, and wire it into your AI tools. Use when the user wants to set up or refresh project memory, or rewire it into a tool. Not for editing one existing memory file.
10-learn
Capture durable project learnings. Use when the user wants to remember, record, or formalize a decision, convention, lesson, pitfall, reusable workflow, or review finding. Not for preferences or temporary notes.
pm-crystallize
Promote a project-local lesson to /.great-pm/decisions.md (cross-project memory) when it has 3+ hits with high confidence. Surfaces candidates as PROPOSALS — never auto-promotes; human approval required.
pm-learn
Auto-extract repeatable patterns, decisions, and cost outliers from the current session and append to .great-pm/lessons.md. Promotes ≥3-occurrence patterns to /.great-pm/decisions.md. Runs continuous-learner.
ima-skill
A general set of instructions for using a personal knowledge base and notes service to store, search, browse, and retrieve information.