rohitg00/agentmemory
16 skills117k installs425k starsGitHub
Install
npx skills add https://github.com/rohitg00/agentmemorySkills in this repo
1RememberA skill that persists information to agentmemory's long-term storage with concept-based tagging for future retrieval. Developers invoke it when users say 'remember this', 'save this', 'note that', or want to preserve knowledge across sessions. The workflow extracts core insights from user input, generates 2-5 specific lowercased concept tags (like 'jwt-refresh-rotation' rather than generic 'auth'), captures referenced file paths, and calls memory_save to persist the data. The skill emphasizes preserving original user phrasing and creating retrievable tags so future recall operations can find the stored information. It pairs with recall for retrieval and forget for deletion, forming a complete memory management system for AI agents.9.2kinstalls2Session HistorySession History displays recent project sessions as a clean timeline using the memory_sessions tool. Developers invoke it when users ask "what did we do last time" or request an overview of previous work. The skill calls memory_sessions with a limit of 20, then presents results in reverse chronological order showing session ID, project name, timestamp, status, and observation counts. For sessions with observations, it surfaces key highlights by type and title, along with session summaries and decisions when available.9.1kinstalls3RecallThe recall skill searches agentmemory using hybrid BM25 plus vector plus graph search to retrieve past observations, sessions, and learnings. Developers invoke it when users ask 'recall', 'what did we do about', 'did we ever', or need context from previous sessions. It calls memory_smart_search with a query and limit, returning results grouped by session with importance scores and observation types (decision, code, etc). High-importance observations (>= 7) are surfaced first. When no results are found, it suggests 2-3 alternative search terms instead of fabricating context.9.1kinstalls4ForgetA governance skill that safely removes data from agentmemory storage through a two-step confirmation workflow. Developers invoke it when users explicitly request memory deletion for privacy or cleanup. The skill first searches agentmemory using memory_smart_search to locate matching observations, displays results to the user for review, waits for explicit confirmation, then executes memory_governance_delete with specific memory IDs. It prevents accidental data loss by requiring explicit yes confirmation and showing exactly what will be deleted before proceeding. The workflow collects individual memory IDs from search results rather than accepting bare session identifiers.9.1kinstalls5Commit ContextThis skill connects code locations to their originating agent sessions by combining git blame/log with session memory lookups. Developers invoke it when they need to understand why specific code exists, what the agent was doing during that commit, or who authored a change. The workflow uses git blame for line ranges, git log for functions or files, then queries memory_commit_lookup with the SHA to retrieve linked session details including observations, summaries, and timestamps. When commits predate session linking (commit: null), it reports that fact without inventing context.9kinstalls6RecapRecap generates time-windowed summaries of agent sessions for the current project, grouping completed work by calendar date with observation highlights. Developers invoke it when asking "what have we been doing" or "recap this week" to review recent progress without reading full session logs. The workflow queries memory_sessions with a time filter (today, last N days, this week), filters results to the current working directory, then pulls high-importance observations (score >=7) via memory_recall. Output shows session IDs, titles, observation counts, and indented highlights, ending with aggregate counts across the window.9kinstalls7HandoffResume the most recent agent session for the current working directory, leading with any unanswered question. Use when the user says "where were we", "resume", "handoff", "pick up where I left off", or starts a session with no fresh context. --- name: handoff description: Resume the most recent agent session for the current working directory, leading with any unanswered question. Use when the user says "where were we", "resume", "handoff", "pick up where I left off", or starts a session with no fresh context. argument-hint: "[optional cwd override]" user-invocable: true --- The user wants to resume work. Optional cwd override: $ARGUMENTS ## Quick start ```json memory_sessions { "limit": 20 } ``` Pick the most recent session whose `cwd` matches this project, then: `memory_recall { "query": "<session top concepts>", "limit": 10 }`. Expected output: ```text Resuming 7f3a9c2 "Auth refresh rework". Open question: should logout revoke all device tokens or just the current one?9kinstalls8Commit HistoryList recent git commits linked to agent sessions, optionally filtered by branch or repo. Use when the user asks "show agent commits", "what has the agent shipped", "list linked commits", or wants commits with their session context. --- name: commit-history description: List recent git commits linked to agent sessions, optionally filtered by branch or repo. Use when the user asks "show agent commits", "what has the agent shipped", "list linked commits", or wants commits with their session context. limit=...]" user-invocable: true --- The user wants a list of agent-linked commits. Filter args: $ARGUMENTS ## Quick start ```json memory_commits { "branch": "main", "limit": 20 } ``` Expected output: ```text 9a1b2c3 main 2026-06-07 "rotate refresh tokens" · session 7f3a9c2 (14 obs) b21d004 main 2026-06-05 "rate limiter audit" · session b21d004 (9 obs) ``` ## Why Render only the commits the tool returned, newest first. An empty result means the filter matched nothing, not that work is missing.9kinstalls9Agentmemory AgentsThe agentmemory-agents skill explains how agentmemory wires into host coding agents through agentmemory connect. REST is the underlying protocol; MCP-only hosts use the stdio MCP bridge adapter. Quick start runs agentmemory connect for claude-code, cursor, codex, gemini-cli, and other supported names listed in generated REFERENCE.md. After wiring, restart the host or run MCP reload so the full tool set appears; only seven tools usually means the MCP shim cannot reach the server. The workflow detects the calling agent, defaults to claude-code when unknown, runs connect with a supported adapter name, and verifies tool visibility. Action skills like remember and recall install separately via npx skills add; connect exposes tools while skills teach when to invoke them. Windows requires WSL2 because native Windows runs the server but connect is unsupported. Cross-links cover mcp-tools, rest-api, hooks, and shared troubleshooting guidance for missing observations.6.5kinstalls10Agentmemory Mcp ToolsThe agentmemory-mcp-tools skill indexes every agentmemory MCP tool with parameters and selection guidance. Quick start saves with memory_save using content, concepts, and files, then recalls via memory_smart_search for hybrid BM25, vector, and graph-expanded search. Tool families span capture, retrieve, sessions, commits, lessons, graph queries, structured slots, and governance utilities like memory_audit and memory_heal. Workflow guidance prefers memory_smart_search for open recall, memory_recall for focused queries, and memory_sessions for listings. Agents must read generated REFERENCE.md for exact parameter names because REST handlers whitelist fields and drop unknown keys. Cross-links cover rest-api, config, and user-invocable action skills including remember, recall, recap, and handoff that wrap the most common operations for day-to-day agent memory tasks in coding workflows. Agents should read SKILL.md quick start steps, verify required binaries and environment variables, and follow reference files for exact parameters before calling tools.6.4kinstalls11Agentmemory ConfigThe agentmemory-config skill agentmemory configuration, environment variables, ports, and feature flags. Use when enabling a feature, changing ports, setting an API key, configuring auth, or explaining why a feature is off by default.. agentmemory reads configuration from the environment and from `~/.agentmemory/.env` (one `KEY=value` per line, no `export` prefix). Restart the server after changing it. Quick start Enable richer memory and set a provider key in `~/.agentmemory/.env`: Defaults worth knowing - No API key is required. Without one, agentmemory runs zero-LLM with BM25 plus local embeddings. - Token-spending features ship OFF on purpose: `AGENTMEMORY_AUTO_COMPRESS` (LLM summaries) and `AGENTMEMORY_INJECT_CONTEXT` (auto context injection) both cost tokens proportional to tool-use frequency. - Tool visibility: `AGENTMEMORY_TOOLS=all` (default) or `core` for the lean set. - Auth: set `AGENTMEMORY_SECRET` Agents should read SKILL.md quick start steps, verify required binaries and environment variables, and follow reference files for exact parameters before calling tools.6.4kinstalls12Agentmemory Rest ApiThe agentmemory-rest-api skill The agentmemory HTTP REST API surface, the primary protocol for talking to the memory server. Use when calling agentmemory over HTTP, when MCP is unavailable and you need a fallback, or when integrating a host that does not speak MCP.. REST is agentmemory's primary surface. MCP is a bridge on top of it. Every memory operation has an HTTP endpoint under `http://localhost:3111/agentmemory/*`. Quick start Auth By default localhost is open and no auth is needed. When `AGENTMEMORY_SECRET` is set, every request needs `Authorization: Bearer $AGENTMEMORY_SECRET`. See agentmemory-config. Conventions - Save returns `201`, reads return `200`, validation errors return `400`. - Handlers whitelist body fields and drop unknown ones, so passing extra keys is safe but ignored. - The port is configurable with `--port` or `--instance`; streams, viewer, and engine derive6.4kinstalls13Agentmemory HooksThe agentmemory-hooks skill The agentmemory plugin hooks that capture observations automatically across the agent session lifecycle. Use when explaining how memory gets captured without manual saves, when debugging missing observations, or when tuning what gets recorded.. The Claude Code plugin registers lifecycle hooks so memory is captured automatically. You do not have to call `memory_save` for routine work; the hooks observe tool use, prompts, and session boundaries and write observations for you. Quick start Install the plugin and the hooks register themselves: Watch observations land live at `http://localhost:3113`. What the hooks do - Session start and end frame each unit of work and let `handoff` resume it. - Tool-use hooks capture what changed and why, the raw material for `recall` and `recap`. - Prompt-submit captures intent. Agents should read SKILL.md quick start steps, verify required binaries and environment variables, and follow reference files for exact parameters before calling tools.6.4kinstalls14Agentmemory ArchitectureThe agentmemory-architecture skill How agentmemory is built, the iii engine primitives it runs on, its storage model, ports, and the viewer. Use when reasoning about how memory is stored or retrieved end to end, when extending the system, or when answering how agentmemory works under the hood.. agentmemory is a memory server for coding agents. It runs locally, captures observations, indexes them for hybrid retrieval, and serves them back over REST and MCP. It is built on the iii engine. iii primitives Everything is a function, a trigger, or worker state on the iii engine. There is no separate plugin system; the worker registers functions (`mem::*`) and HTTP triggers (`api::*`) and the engine routes calls. agentmemory does not bypass iii; new capability is a new function plus a trigger. Retrieval model Recall is hybrid: BM25 keyword search6.4kinstalls15Write Agentmemory SkillThe write-agentmemory-skill skill The house format and rules for writing or updating an agentmemory skill. Use when adding a new skill, restructuring an existing one, or reviewing a skill contribution for consistency.. agentmemory skills follow one tiered format so they stay skimmable, accurate, and current. Match it exactly. Directory layout SKILL.md rules - Frontmatter: `name`, `description`, optional `argument-hint`, and `user-invocable`. Set `user-invocable: true` only for skills the user runs as a slash command; reference and knowledge skills are `false`. - Description is two sentences and the only thing the agent sees when deciding to load the skill. Sentence one states the capability. Sentence two starts "Use when" and lists concrete triggers. Keep it distinct from sibling skills, under 1024 chars, third person. Agents should read SKILL.md quick start steps, verify required binaries and environment variables, and follow reference files for exact parameters before calling tools.6.4kinstalls16Agentmemoryagentmemory is a Claude Code skill in the AI & Agent Building category. Persistent memory for AI coding agents -- captures tool usage, compresses via LLM, injects context into future sessions0installs