chunk-retrospective

chunk-retrospective is an agent for Claude Code from MartyBonacci/specswarm. It costs 129 tokens per session (2,653 once invoked), scanned A, original, MIT.

An agent that turns the completed work on one feature into a small set of reusable lessons stored as project memory.

In plain words
What is it for?
Use it near feature completion to record lessons for future planning, implementation, and pre-flight checks.
Why use it?
It prevents decisions, mistakes, and corrections from being lost when a coding session ends.

Agent for Claude Code

Written for Claude Code: effort in frontmatter. Also seen: model in frontmatter; names the AskUserQuestion tool; mentions Claude Code.

Part of the ss plugin — 10 skills, 34 commands, 6 agents, 4 hooks shipped together

Install

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.

agentmods
npx agentmods add agents/martybonacci/specswarm/chunk-retrospective
Clone the repo
git clone --depth 1 https://github.com/MartyBonacci/specswarm

Made for: Claude Code.

Or install ss, the plugin that ships this one along with the rest of its 10 skills, 34 commands, 6 agents, 4 hooks.

Wrote 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.

agentmods badge for chunk-retrospective

README.md
[![agentmods](https://agentmods.dev/badge/agents/martybonacci/specswarm/chunk-retrospective.svg)](https://agentmods.dev/agents/martybonacci/specswarm/chunk-retrospective)
Your own site
<a href="https://agentmods.dev/agents/martybonacci/specswarm/chunk-retrospective"><img src="https://agentmods.dev/badge/agents/martybonacci/specswarm/chunk-retrospective.svg" alt="Measured on agentmods" height="20"></a>
Per session 129 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 2,653 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00129 $0.02653
Opus 5 $0.00064 $0.01326
Sonnet 5 $0.00026 $0.00531
Haiku 4.5 $0.00013 $0.00265

Measured 6d ago against content hash fd089a968533, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-06, from the pricing page.

Security

Grade A, and why

chunk-retrospective 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 6d 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.

plugins/ss/agents/chunk-retrospective.md · 181 lines

How it starts

The opening of the file, as written. The whole thing — 181 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are the SpecSwarm Chunk Retrospective agent — invoked once per completed feature/chunk to convert session-bound lessons into durable, classifiable memory files.

Model rationale (v7.7.0): This agent runs on opus. Synthesizing lessons from raw signals requires meta-cognition + classification + concise writing — all judgment-heavy. Memory entries written here compound across chunks, so quality matters more than speed.

Mission

The fundamental problem: a long-running Claude Code session accumulates lessons across many edits, decisions, drifts, and corrections. When the session ends, that knowledge evaporates. Memory files persist; session history does not.

Your job: distill the chunk into 1–3 durable memory entries, written directly via your Write tool, that the next chunk's spec-mentor, builder, or /ss:preflight checks will read and benefit from.

Input you receive

Your invoking command (/ss:retrospective) passes a structured context bundle containing these fields. Read them carefully before writing anything.

  • feature_id — e.g., 002-database-schema-migrations-seeds
  • feature_dir — absolute path to .specswarm/features/NNN-name/
  • parent_branch — git branch this feature diverged from (e.g., main)
  • commitsgit log <parent>..HEAD output: short hash + subject + body, one commit per record. Reflects real decisions and corrections made during implementation.
  • tasks_md_path — absolute path to the feature's final tasks.md
  • verified_tasks — list of task IDs that reached .verified state (PASS)
  • flagged_tasks — list of task IDs that reached .flagged state with their details field (DRIFT / NEEDS-MARTY)
  • recent_interventions — paths to intervention_*.md files captured during this chunk
  • memory_dir — absolute path where you'll write new memory files
  • memory_index_path — absolute path to MEMORY.md (if present)
  • existing_memory_summary — short list of existing memory file names + their description: lines, for dedup

Read the full file on GitHub · 181 lines

Changes

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.

  1. 6d ago First seen · 181 lines · 129 tokens per session scan A fd089a968533

Subscribe to this mod's changes

chunk-retrospective is an agent published in the GitHub repository MartyBonacci/specswarm (65 stars, last pushed 1mo ago), licensed MIT. It adds 129 tokens to every session and 2,653 once invoked, about $0.0006 per session on Opus 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.

Related

Other agents, from other repositories

context-manager

Use this agent when you need to manage context across multiple agents and long-running tasks, especially for projects exceeding 10k tokens. This agent is essential for coordinating complex multi-agent workflows, preserving context across sessions, and ensuring coherent state management throughout extended development…

czlonkowski/n8n-mcp · 0 tokens

starlight-repos-extractor

Tier: Phase 1 extractor Dispatched via: Agent tool Output contract: JSONL atoms appended to.

frankxai/Starlight-Intelligence-System · 7 tokens

portable-memory-parent-orchestrator

Top orchestrator for the portable-process-memory feature. Delegates to sync-transport (push/fetch folded into the verbs, plain-git, credential inheritance, offline-fail-safe) and event-fold (ownership events + the fail-closed divergence tripwire in the gate fold). Architect-only; coordinates portability/transport work…

seanrreid/RAD_framework · 77 tokens

context

Agent "context" from hannsxpeter/godpowers, covering scope, context, decisions, rules and workflows.

hannsxpeter/godpowers · 0 tokens

context-finder

Read-only, memory- and index-aware codebase search. Use for any investigation — "where is X", "how does Y work", "what calls Z", "is W still used", "where is V configured", "does this event/pattern get emitted anywhere" — BEFORE reaching for grep. Consults the knowledge graph, code index, and prior session memory…

futuregerald/futuregerald-claude-plugin · 111 tokens

wiki-ingest

Use this agent when ingesting URLs, files, or pasted text into the vault during automated maintenance cycles. Typical triggers include dev-loop IDLE DISCOVERY ingestion, batch source processing, or converting raw captures to typed-knowledge pages. See "When to invoke" in the agent body for worked scenarios.

karlorz/llm-wiki · 64 tokens