New: 98,000+ skills indexed from GitHub repos with 100+ stars · every skill safety-checkedBrowse top skillsTop skills
Esc to closeBrowse all skills
Submit a skillBrowse skills
Submit a skillBrowse skills

Agent skill

TMA1 Session Search

by tma1-ai in tma1-ai/tma1

Searches earlier agent sessions on this project by keyword and reads back a session's conversation, answering with quotes from what was actually said.

Apache-2.0Auto-check passedAgent Workflows

Install TMA1 Session Search

skills CLI
$ npx skills add tma1-ai/tma1 --skill tma1-search -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install tma1-ai/tma1 tma1-search --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/tma1-ai/tma1.git skills-src && mkdir -p .claude/skills && cp -r skills-src/claude-plugin/skills/tma1-search .claude/skills/tma1-search && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
tma1-search
GitHub stars
119
Token cost
~904 tokens
SKILL.md length
373 words
Files
1
Skills in repo
5
Repo updated
First seen
Licence
Apache-2.0

At a glance

Searches earlier agent sessions on this project by keyword and reads back a session's conversation, answering with quotes from what was actually said.

  • Works in 4 steps: Search — mcptma1search_sessions with → Report the matching sessions with their… → Read — mcptma1get_session_transcript… → …
  • Finding how a problem was fixed in an earlier session
  • SKILL.md covers Two entry points, Workflow, Reading the search response and Search behaviour worth knowing, plus 1 more section
  • Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md

What it does

TMA1 keeps every agent session on the machine in a local database, and this skill is the read path. A command such as /tma1-search retry backoff searches the current project over the last 7 days. Flags narrow the search to one agent, widen the day window or cover all projects, and --session reads a particular session by a full ID or a prefix of at least 6 characters. A bare token is never assumed to be a session ID, because commit hashes look the same.

The workflow is to search with mcp__tma1__search_sessions, report the matching sessions with snippets so you can choose, read with mcp__tma1__get_session_transcript while following next_offset when more messages exist, and then answer with quotes from the earlier session rather than a summary. It explains the two match modes, an indexed whole-word, case-sensitive pass and a wider substring fallback that may add noise, warns when the scan was truncated, and says plainly when nothing matched instead of inventing a past session.

When your agent uses it

  • Finding how a problem was fixed in an earlier session
  • Recalling what was decided about a topic last week
  • Reading back a session you were given an ID for
  • Looking only at what Codex or another agent tried

Example prompts

  • “What did we decide about retry backoff? Search my sessions.”
  • “Find the session where the flaky test was fixed and quote the fix.”
  • “Search codex sessions only for migration, going back 30 days.”

Requirements

  • The TMA1 plugin with its search_sessions and get_session_transcript MCP tools
  • Pre-approved tools (allowed-tools): mcp__tma1__search_sessions, mcp__tma1__get_session_transcript

Workflow steps

4 steps, taken from the first numbered list in SKILL.md.

  1. Search — mcptma1search_sessions with
  2. Report the matching sessions with their snippets, so the user can
  3. Read — mcptma1get_session_transcript with the session_id
  4. Answer with quotes. The value here is the earlier session's actual

What it can do on your machine

Read from SKILL.md and the folder at commit f3dfed8. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves these tools, so the agent can use them without asking each time:

    • mcp__tma1__search_sessions
    • mcp__tma1__get_session_transcript

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    No scripts in the folder and no shell commands in SKILL.md.

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

TMA1 Session Search loads about 904 tokens when it runs. Until then it costs about 120 tokens; SKILL.md has 373 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~120
When it runs · the whole SKILL.md, loaded when a task matches
~904

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from tma1-ai/tma1 at commit f3dfed8, republished under its Apache-2.0 licence (© tma1-ai). 373 words, ~904 tokens.

Download SKILL.mdSave it as .claude/skills/tma1-search/SKILL.md (or your agent's skills folder).
name
tma1-search
description
Search past agent sessions on this project by keyword, and read back any session's conversation. Invoke when the answer lives in an earlier session — how something was fixed before, what was decided, what an agent tried last week — or when the user hands you a session id. Trigger phrases: "what did we decide about", "how did we fix this before", "search my sessions", "find that session", "read session <id>", "之前是怎么解决的", "上次讨论过", "找一下以前的 session", "/tma1-search".
allowed-tools
mcp__tma1__search_sessions, mcp__tma1__get_session_transcript
argument-hint
<keyword> [--agent X] [--days N] [--all-projects] | --session <id>

TMA1 keeps every agent session on this machine — yours and other agents' — in a local database. This skill is the read path: find the session, then read it.

Two entry points

/tma1-search retry backoff            # keyword search, this project, last 7 days
/tma1-search flaky test --agent codex # only Codex's sessions
/tma1-search migration --days 30      # widen the window
/tma1-search pipeline --all-projects  # every project, not just this one
/tma1-search --session 3d8f1d0a       # read one session directly

--session takes a full id or a prefix of 6+ characters — the abbreviated id in the <tma1-context> block works. Do not infer a session id from a bare token: commit hashes look the same, and oc: / cp: ids don't look like hex at all. Anything that isn't behind --session is search text.

Workflow

  1. Search — mcp__tma1__search_sessions with:
    • query: the keyword text (everything that isn't a flag)
    • agent_source: --agent value if given (codex / openclaw / copilot / claude / self); omit for every agent
    • since_min: --days N × 1440, default omitted (7 days)
    • project: "*" when --all-projects is given; otherwise omit
  2. Report the matching sessions with their snippets, so the user can pick. One obvious hit and a clear question → go straight to step 3.
  3. Read — mcp__tma1__get_session_transcript with the session_id from step 1. has_more: true means older messages exist; call again with the returned next_offset if the answer isn't in this page.
  4. Answer with quotes. The value here is the earlier session's actual words, not your summary of them.
Show full SKILL.md (173 more words)Show less

Reading the search response

  • sessions[] — session_id, agent_source, last_activity_ago, hit_count, and snippets[] (each a window around a match).
  • match_mode — term is the indexed pass: whole words, case-sensitive. substring means that pass found nothing, so the search widened to a case-folded substring scan; results may include noise, say so.
  • scope_truncated: true — this project has more sessions in the window than the search scanned. Narrow with --agent or --days.
  • note — present only when nothing matched. Say so plainly; don't invent a plausible past session.

Search behaviour worth knowing

The indexed pass matches whole words and is case-sensitive, so Cwd does not find peerCwdFilter and neither does peercwdfilter. Search for the identifier as it is written in the code. When that pass finds nothing the search widens to a case-folded substring scan and tells you so with match_mode: substring — that is the mode that rescues a mistyped case, at the cost of noisier hits.

  • Browsing by agent rather than keyword → /tma1-peer.
  • Aggregates over the raw tables (cost by model, tool failure rates) → the tma1 skill's exec_query.

© tma1-ai, Apache-2.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

Just SKILL.md in claude-plugin/skills/tma1-search of tma1-ai/tma1.

Open the folder on GitHubat commit f3dfed8

Compare with similar skills

TMA1 Session Search next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

TMA1 Session Search compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
TMA1 Session Search this skilltma1-ai/tma1119—~904Automated safety check: PassApache-2.0
Memori MCP Memory UsageMemoriLabs/Memori17k—~3.8kAutomated safety check: PassMIT
Ourmemourmem/omem1741 repos~3.4kAutomated safety check: PassCustom licence
Claude Code Auto-Capture HookNateBJones-Projects/OB14.7k—~1.3kAutomated safety check: NotesCustom licence
Memory Searchthedotmack/claude-mem99k—~1.4kAutomated safety check: PassApache-2.0
Engram MemoryPatdolitse/piia-engram163—~1.3kAutomated safety check: PassAGPL-3.0-or-later

Similar skills

  • Memori MCP Memory Usage

    MemoriLabs/Memori

    Teaches an MCP-connected agent when and how to call Memori's recall, summary, compaction, augmentation, feedback and quota tools to keep context across sessions.

    17k GitHub stars~3.8k tokensUpdated 8 days ago
    Agent WorkflowsAuto-check passed

  • Ourmem

    ourmem/omem

    Shared memory that never forgets. An agent skill from ourmem/omem.

    174 GitHub starsUsed in 1 repo~3.4k tokens
    Agent WorkflowsAuto-check passed

  • Claude Code Auto-Capture Hook

    NateBJones-Projects/OB1

    Fires a Stop hook that captures a Claude Code session transcript automatically when the session ends without a verbal wrap-up.

    4.7k GitHub stars~1.3k tokensUpdated 2 days ago
    Agent WorkflowsAuto-check: notes

  • Memory Search

    thedotmack/claude-mem

    Searches the claude-mem database of past sessions in layers (index, timeline, full details) to find how a problem was solved before.

    99k GitHub stars~1.4k tokensUpdated 2 days ago
    Agent WorkflowsAuto-check passed

  • Engram Memory

    Patdolitse/piia-engram

    Routes continuity and recall requests to Engram, a local-first MCP memory and identity layer that saves user-approved lessons, decisions and playbooks.

    163 GitHub stars~1.3k tokensUpdated today
    Agent WorkflowsAuto-check passed

  • Compartment Session Sweep

    MaxFreedomPollard/Compartment

    Sweeps a conversation before compaction and saves durable facts, decisions and session records into the Compartment encrypted memory vault as short one-claim memories.

    579 GitHub stars~750 tokensUpdated 11 days ago
    Agent WorkflowsAuto-check passed

More from tma1-ai/tma1

  • TMA1 Peer Sessions

    tma1-ai/tma1

    Lists recent sessions on the current project from other agents such as Claude Code, OpenClaw and Copilot CLI, or from your own earlier work.

    119 GitHub stars~1.3k tokensUpdated 2 days ago
    Auto-check passed

  • TMA1 Peer Sessions

    tma1-ai/tma1

    Lists recent sessions of peer coding agents such as Codex, OpenClaw or Copilot CLI, or your own past sessions, so you can act on their work without copying between terminals.

    119 GitHub stars~1.1k tokensUpdated 2 days ago
    Auto-check passed

  • TMA1 Session Search

    tma1-ai/tma1

    Searches past agent sessions on a project by keyword through the TMA1 MCP server and reads back a chosen session so answers can quote it directly.

    119 GitHub stars~766 tokensUpdated 2 days ago
    Auto-check passed

  • TMA1 Observability Query

    tma1-ai/tma1

    Answers questions about agent spend, token use, traces, events, errors and tool usage by running read-only SQL against a local TMA1 observability store.

    119 GitHub stars~5.1k tokensUpdated 2 days ago
    Auto-check: notes

Works with

Categories

Questions about TMA1 Session Search

What does TMA1 Session Search do?

Searches earlier agent sessions on this project by keyword and reads back a session's conversation, answering with quotes from what was actually said. TMA1 keeps every agent session on the machine in a local database, and this skill is the read path. A command such as /tma1-search retry backoff searches the current project over the last 7 days.

When should I use TMA1 Session Search?

TMA1 Session Search fits situations like: finding how a problem was fixed in an earlier session; recalling what was decided about a topic last week; reading back a session you were given an ID for; looking only at what Codex or another agent tried.

How do I install TMA1 Session Search in Claude Code?

Run `npx skills add tma1-ai/tma1 --skill tma1-search -a claude-code`. Or copy the skill folder (claude-plugin/skills/tma1-search in tma1-ai/tma1) into .claude/skills/tma1-search in your project. Claude Code loads it when a task matches its description.

How do I install TMA1 Session Search in Codex?

Run `npx skills add tma1-ai/tma1 --skill tma1-search -a codex`. Or copy the skill folder (claude-plugin/skills/tma1-search in tma1-ai/tma1) into .agents/skills/tma1-search in your project. Codex loads it when a task matches its description.

Can I use TMA1 Session Search in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add tma1-ai/tma1 --skill tma1-search -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/tma1-search, .gemini/skills/tma1-search, .github/skills/tma1-search and .opencode/skills/tma1-search in your project.

What does TMA1 Session Search need to run?

SKILL.md names no scripts, command-line tools or credentials: TMA1 Session Search is instructions for the agent only. Our summary lists: The TMA1 plugin with its search_sessions and get_session_transcript MCP tools. Its frontmatter pre-approves these tools: mcp__tma1__search_sessions, mcp__tma1__get_session_transcript.

Does TMA1 Session Search access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is TMA1 Session Search safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does TMA1 Session Search use?

TMA1 Session Search is published under the Apache-2.0 licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does TMA1 Session Search use?

About 904 tokens (SKILL.md is roughly 3.6k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to TMA1 Session Search?

Skills that share tags, products or a category with TMA1 Session Search: Memori MCP Memory Usage (MemoriLabs/Memori, 17k stars), Ourmem (ourmem/omem, 174 stars), Claude Code Auto-Capture Hook (NateBJones-Projects/OB1, 4.7k stars) and Memory Search (thedotmack/claude-mem, 99k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains TMA1 Session Search?

tma1-ai (a GitHub organization) maintains it in tma1-ai/tma1, which has 119 GitHub stars. The repository holds 5 skills in this directory. The repository was last updated on October 8, 2026.

Source: tma1-ai/tma1 on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.