Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/manishiitg/coding-agent-loopnpx agentmods add skills/manishiitg/coding-agent-loop/recall-conversationsWrote 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/manishiitg/coding-agent-loop/recall-conversations)<a href="https://agentmods.dev/skills/manishiitg/coding-agent-loop/recall-conversations"><img src="https://agentmods.dev/badge/skills/manishiitg/coding-agent-loop/recall-conversations/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/manishiitg/coding-agent-loop/recall-conversations"><img src="https://agentmods.dev/badge/skills/manishiitg/coding-agent-loop/recall-conversations.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 | $0.00040 | $0.00890 |
| Opus 5.5 | $0.00016 | $0.00356 |
| Sonnet 5.5 | $0.00008 | $0.00178 |
| Haiku 4.5 | $0.00004 | $0.00089 |
Grade A, and why
recall-conversations scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 29d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 80 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Recall an earlier conversation
Every conversation is logged by the platform, one JSON file per conversation, under the family's chat-history folder — two levels up from your own folder:
../../chat_history/<yyyy-mm-dd>/session-<session-id>-conversation.json
The date folder is the day the conversation started; a conversation that ran
over several days stays in its first day's folder. updated_at inside the file
is its last activity. Newest first is ls -t ../../chat_history/*/session-*-conversation.json.
Which file is which
- An activity's conversation (the child and the tutor): the activity's
product.jsonholds itssession_id:
An activity from before September 2026 keeps its older log assid=$(python3 -c "import json,sys; print(json.load(open(sys.argv[1]))['session_id'])" "activities/<slug>/product.json") ls ../../chat_history/*/session-${sid}-conversation.jsonlegacy-conversation.jsoninside the activity folder instead (a plainmessageslist) — read that one directly. - The parent's earlier chats with you: the files whose
runtime.workspace_pathends inChats/SparkQuill(an activity's ends inactivities/<slug>). The current chat is already in front of you — its session id is thesparkquill/mainentry of../../chat_history/product-conversations.json; skip that file, the earlier ones are the previous chats. - WhatsApp turns run inside these same conversations; there is no separate log.
Reading a log
A log is two files: session-<id>-conversation.json (the record, whose
conversation_history holds only the latest turns when history_tail is true)
and session-<id>-conversation.history.jsonl beside it (every turn, one JSON
object per line). Read the .history.jsonl when it exists; an older log has no
.history.jsonl and keeps every turn in conversation_history.
Each turn is {"Role": ..., "Parts": [...]}.
Role is human (the parent, or the child in an activity), ai (you, or the
tutor), tool (a tool result) or system. Only text parts matter for recall:
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.
- 29d ago Changed · +11 lines c15c5e5506ad
- 1mo ago First seen · 69 lines · 40 tokens per session scan A 7ef09dce5bd2
recall-conversations is a skill published in the GitHub repository manishiitg/coding-agent-loop (13 stars, last pushed 27d ago), licensed MIT. It adds 40 tokens to every session and 890 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 2026-09-07.
Other skills, from other repositories
semantix
Install and use the semantix memory kernel as a middleware in your agent: extract user preferences / workflows / experience from past sessions, retrieve and inject them on demand. One binary + your agent's own tools.
distill-session-knowledge
Offline-mine this project's pi session JSONL logs into reusable, verified knowledge: extracts faults, decisions, corrections, procedures and docs, promotes only recurring patterns, and routes artifacts into skillmanage, memory and docs. Use on "mine my sessions", "distill session knowledge", "extract lessons from…
skills-constitution
A set of rules that makes an agent check its memory and available skills before handling professional tasks such as coding, file work, data analysis, or deployment.
goga-cell-go
Golang rules for implementing CODEMANIFEST contracts.
recall
Use before assuming you're the first to hit something in an octomux loop/scheduled task — pull what past runs learned about the thing you're touching via octomux recall --query.
design-memory
Use at project start and completion to OBSERVE and record how the user designs — the decisions they make, the styles they reach for, their habits and inclinations across projects. This is a descriptive record (a mirror), NOT a controller — it is never fed back to steer the work. Per-project direction comes from…