| name | project-retrospective |
| description | Use when a project needs multi-session retrospective analysis — after milestones, before starting a new phase, when vision drift is suspected, or every 3-5 sessions. Triggers: user says "retro", "run a retro", "since-last-retro"; asks to analyze patterns, recurring mistakes, or correction trends across multiple sessions; says "how did we get here" or "project history" in a project-arc context (not single-file diffs or README edits); or notes that project memory has drifted from what actually happened across sessions. NOT for: single-session review, memory cleanup, retro templates, or git-level file diffs. |
| argument-hint | ["full | since-last-retro | last-N"] |
Project Retrospective
Hermes adaptation of the claudskills.com community skill (same name). Four
portability changes, methodology unchanged: (1) session source is the Hermes
session store (session_search) instead of ~/.claude/exports/; (2) historian
dispatch uses Hermes delegate_task instead of the Claude Code Agent tool;
(3) rules/memory targets are Hermes-native (project memory files, the memory
tool, skills) instead of ~/.claude/rules/; (4) all model-specific wording
("Claude", "opus", "Haiku/sonnet") is generalized to "the assistant".
Analyze a project's session history by dispatching parallel historian agents to read each session record, then synthesizing their findings into a structured analysis document. The value is in the extraction criteria — domain-specific signals, not generic summarization.
Modes
Two modes, differentiated by output format (not historian cost):
| Mode | Output | When to Use |
|---|
| full | Standalone ANALYSIS (superset, safe to delete prior) | Phase transitions, major incidents, no prior retro exists, reset the chain |
| incremental | Delta UPDATE (references prior retro, never delete prior) | Periodic check-ins (every 3-5 sessions), ongoing projects with an existing retro baseline |
Both modes reuse prior historian work when a prior retro exists — only NEW session records get fresh historians. The difference is what the synthesizer produces: a standalone document vs. a delta document.
Arguments
The mode/scope argument comes from the user's request wording: "full", "since-last-retro", or "last-N" (e.g. "retro last 5"). If the user gives no argument, apply the context-aware default below.
| Argument | Behavior |
|---|
| (none) | Context-aware default. Check for prior retros first (Phase 1.5). If a prior retro exists, default to incremental. If no prior retro exists, default to full. |
full | Explicit full mode. User override — produce a standalone ANALYSIS. Still reuse prior historian work (only dispatch historians for new session records). |
since-last-retro | Explicit incremental mode. Produce a delta UPDATE against the prior retro. |
last-N | Full mode scoped to the N most recent session records only. Always dispatches fresh historians for all N session records — does not reuse prior retro content. Skip Phase 1.5 entirely. Use for focused recent-session analysis. |
If the argument doesn't match any of the above, echo it back and ask what was meant.
User override is final. If the user explicitly says full, produce a full ANALYSIS — don't argue or suggest incremental. The context-aware default only applies when no argument is given.
Phase 1: Discover & Validate Session Records
Obtain Hermes session records (first available wins):
session_search retrieval — query the session store by project keywords; take the full message text of each relevant session as one record block, sorted by date.
- User-provided exports — if the user supplies session export files (.txt/.md), read those.
- Hermes CLI export — if a
hermes sessions-style command can dump session text, use it; otherwise fall back to option 1.
If last-N was provided, take only the last N records.
No session records available: Stop immediately. Report the issue and suggest confirming that session history exists (Hermes records sessions automatically in the local store; search by project name or keywords). Do NOT proceed with zero records.
Some records missing: List what was found, proceed with available records, and note gaps in the final output.
Phase 1.5: Detect Prior Retro
In both modes, find the most recent full retro document:
docs/retros/*-PROJECT-HISTORY-ANALYSIS.md
Fallback for pre-v2 projects: If docs/retros/ has no matches, also check docs/*-PROJECT-HISTORY-ANALYSIS.md (pre-v2 output path). If found there, move it to docs/retros/ first, then proceed.
If no prior retro exists anywhere:
- Incremental mode: Report this and switch to full mode automatically. Incremental requires a baseline.
- Full mode: Proceed normally — all session records need fresh historians.
If prior retro found:
- Read it fully — it contains the synthesized analysis of previously-analyzed sessions.
- Extract the session range it covers (from the
**Sessions:** <range> header line).
- Determine which session records are NEW (not covered by the prior retro's session range).
- If zero new session records exist since the prior retro: report "nothing new to analyze — prior retro is current" and stop. This applies to both modes — re-synthesizing the same data produces equivalent output.
- Full mode: The prior retro's content serves as pre-computed extraction for already-analyzed sessions (see Phase 2).
Chaining semantics: Incremental always chains from the last full retro (ANALYSIS file), never from a prior incremental update (UPDATE file). This means multiple incremental updates can accumulate between full retros. Each delta is independently interpretable against the same baseline. To reset the chain, run a full retro.
Gaps in prior retro: If the prior full retro noted missing session records (sessions it couldn't analyze), those gaps are permanent unless a new full retro is run. Incremental mode does not backfill gaps — it only analyzes session records newer than the prior retro's session range.
Phase 2: Spawn Historians (Parallel Background Agents)
Incremental mode: Launch one background agent per NEW session record only.
Full mode with prior retro (Phase 1.5 found one): Only dispatch historians for session records NOT covered by the prior retro. The prior retro's content serves as pre-computed extraction for already-analyzed sessions — pass it to the synthesizer in Phase 3 alongside the new historian reports. This gives full-mode superset output with incremental historian cost.
Full mode without prior retro: Launch one background agent per session record (all session records).
Dispatch each historian with Hermes delegate_task (runs in the background by default; one sub-agent per session record):
delegate_task(
goal: "Historian: {SESSION_LABEL} — analyze the session record below per the extraction template; return clean markdown text only, write no files",
context: "<full extraction template> + <session record text (contents of {FILE_PATH})>"
)
Model selection: Hermes sub-agents inherit the parent model. Session records can be long (30-65K tokens) — if the parent model is weak, extraction quality will drop; configure a stronger model for historians in the delegation settings if needed.
Collection: After ALL historians complete, collect their reports and proceed to Phase 3 (for 15+ reports, delegate the synthesis itself to a dedicated sub-agent — see Phase 3).
Historian Extraction Template
Each historian receives this prompt with {FILE_PATH} and {SESSION_LABEL} filled in:
Read the COMPLETE session record at {FILE_PATH}. This is a Hermes session
record from {SESSION_LABEL}.
Session records contain the visible conversation: user messages, assistant
responses, decisions, corrections, and deliverables. Tool calls may be
collapsed or truncated — sub-agent prompts, memory writes, and sub-agent
details may be behind expansions and NOT visible.
Extract what IS visible: user messages, assistant responses, decisions,
corrections, and deliverables.
Extract the following. Include brief quotes or concrete references — not
vague summaries.
1. **How session started**: First user prompt. Continuation or fresh?
How was context established?
2. **Original intent vs actual**: What was planned vs what happened.
Why different?
3. **Key decisions**: Technology, architecture, process decisions — with
rationale and whether they held or were reversed.
4. **User corrections**: Every time the user caught the assistant making a
mistake. QUOTE the correction verbatim. This is the most valuable
data — pattern these by type.
5. **Mistakes and anti-patterns**: What went wrong, root causes,
systemic issues (not just one-offs).
6. **Quality moments**: When user pushed for higher rigor — what was
the ask and what was the assistant's response?
7. **Roadmap evolution**: How did the plan/scope change during the session?
8. **Deliverables**: Files created/modified, PRs opened/merged/closed,
documents produced.
9. **Handoff**: What was stated as the next session's task? Deferred items?
Output as clean markdown. No preamble. Return TEXT DATA only — do NOT
write any files.
Historian failure: Relaunch once. If it fails again, proceed with available reports and note the gap. Do NOT approximate from other historians or from project memory — relaunch, don't guess.
Phase 3: Synthesize
After ALL historians complete, combine reports into a single analysis.
Synthesizer delegation: If fewer than ~15 historian reports, the orchestrator can synthesize inline. For 15+ reports, delegate to a dedicated synthesizer sub-agent (strongest model accessible) — the combined reports plus prior retro content may exceed comfortable inline processing. Pass all historian reports and (if applicable) the prior retro content in the synthesizer's prompt.
Full mode uses the Full Synthesis Template. Incremental mode uses the Incremental Synthesis Template.
Full mode with prior retro input: The synthesizer needs BOTH the new historian reports AND the prior retro content. The prior retro provides the deep extraction for already-analyzed sessions; the new historian reports cover the delta. The synthesizer must treat both as equal primary sources and produce a standalone superset ANALYSIS — not a delta, not a summary of the prior retro with new sections appended. Re-synthesize the full narrative from all available data.
Incremental mode input: The synthesizer needs BOTH the new historian reports AND the prior retro content (read in Phase 1.5). If delegating synthesis to a sub-agent, include the prior retro text in its prompt — the sub-agent doesn't have Phase 1.5 context.
Full Synthesis Template
# {Project Name} — Project History Analysis
**Generated:** YYYY-MM-DD HH:MM UTC | **Sessions:** range | **Agents:** N+1
## 1. Original Vision vs Current State
The user's original vision, how it evolved, honest drift assessment —
were changes justified or accidental?
## 2. Roadmap: Original vs Actual
| Session | Planned | Actual | Deviation Justified? |
## 3. Decision Log
| Session | Decision | Rationale | Status (held/reversed/superseded) |
## 4. Mistakes and Corrections
Pattern mistakes by type. Common assistant behavioral patterns to seed:
- Deferral disguised as process (effort avoidance)
- Overconfidence without verification (asserting not checking)
- Not following own rules (loaded but not consulted)
- Fabricated/unverified claims (training data biases)
Also surface any project-specific patterns beyond these.
Count instances per pattern — reveals systemic issues.
## 5. User Teaching Moments
Recurring lessons the user had to teach. These are the user's priorities
and standards — future sessions must internalize them.
## 6. What's Valid Now
Current state: artifacts, what's on main, what's pending. Clear inventory.
## 7. Recommended Next Step
Justified from FULL history, not just the latest project memory. If history
suggests a different priority than project memory, say so.
Incremental Synthesis Template
# {Project Name} — Project History Update
**Generated:** YYYY-MM-DD HH:MM UTC | **New sessions:** range | **Prior baseline:** prior-retro-filename
**Cumulative sessions:** full-range | **Agents:** N+1
## Prior Retro Summary
One-paragraph summary of what the prior retro established as the
project state, key patterns, and outstanding issues.
## 1. What Changed Since Last Retro
New decisions, direction shifts, scope changes in the delta sessions.
Reference prior retro for context where relevant.
## 2. New Decisions
| Session | Decision | Rationale | Status |
## 3. Correction Pattern Update
Carry forward the prior retro's correction patterns. For each pattern:
- Prior count + new count = cumulative total
- Trend: improving / stable / worsening (compare per-session rate)
- Any NEW patterns not in the prior retro get their own entry.
## 4. New Teaching Moments
Lessons from new sessions only. Note which are genuinely new vs
reinforcement of prior patterns.
## 5. Current State & New Deliverables
Updated inventory: what's on main, what's pending, what changed.
List files created/modified, PRs merged, documents produced in new sessions.
## 6. Recommended Next Step
Justified from cumulative history (prior retro + new sessions).
Cross-session deduplication: A decision in S1 referenced in S3 appears once with both session numbers.
Phase 4: Write & Integrate
Write the analysis
Output to the retros directory (create if needed):
Full mode: docs/retros/YYYY-MM-DD-PROJECT-HISTORY-ANALYSIS.md
Incremental mode: docs/retros/YYYY-MM-DD-PROJECT-HISTORY-UPDATE.md
Update project memory
- Add a reference to the retrospective document.
- Evaluate whether findings warrant new rules or memory:
- Project-specific lessons → project memory file (e.g.
.hermes.md) or PROJECT-CONTEXT.md
- General patterns (cross-project) → update the relevant Hermes skill or global memory (
memory tool)
- Existing-rule violations → note the enforcement gap, don't duplicate
- Threshold: happened twice, or high-severity once.
Commit
Commit the analysis and any memory updates.
Deletion Semantics
| Mode | Prior retro safe to delete? | Why |
|---|
| Full | Yes | Full output is a superset — contains everything the prior retro had plus more. Prior becomes redundant. |
| Incremental | Never | Incremental output is a delta — it references and builds on the prior retro. Deleting the prior loses the deep extraction from earlier sessions permanently. The chain of incremental retros forms a linked history. |
Anti-Patterns
- Single agent for all records — each agent needs full context for one session record. Combining loses depth.
- Compressing historian reports — the synthesizer needs full detail for cross-session patterns.
- Skipping "User corrections" — the most valuable signal. Without it, the retro is generic.
- Writing from memory files only — project memory is derivative. Session records are the primary source.
- Serial processing of all records — historians must run in parallel (one background sub-agent per record); sequential runs lose the parallelism the design depends on.
- Weak model for historians — session records are long (50-200KB scale); weak models miss nuance. Use the strongest model available.
- Summarizing instead of extracting — "summarize the session" produces generic output. The 9-point extraction template IS the skill's value.
- Re-analyzing old sessions in incremental mode — the whole point of incremental is O(delta). If you're spawning historians for sessions the prior retro already covers, you're doing it wrong.
- Fresh historians for already-extracted sessions — if a prior retro exists, its content IS the extraction for those sessions. Dispatching new historians to re-read the same session records wastes sub-agents and produces equivalent output. Only dispatch historians for sessions the prior retro doesn't cover.
- Hardcoding default mode without checking project state — the default (no argument) should check whether prior retros exist before committing to full or incremental. A blind default ignores available information.
- Deleting prior retro after incremental — incremental output is a delta, not a superset. The chain breaks.
- Running incremental without a baseline — if no prior retro exists, switch to full mode. Don't produce a delta with nothing to delta against.