harvest

/harvest -- Sprint analytics and retrospective

The user wants quantitative analysis of the current sprint: claim type distributions, evidence quality, stale claims, and velocity metrics.

Arguments

$ARGUMENTS

Optional: --full to generate an HTML retrospective report. --calibrate to run prediction scoring.

Instructions

1. Detect the sprint directory:
2. Run wheat_status to get the current sprint metadata.
3. Identify the sprint directory from the status output (the directory containing claims.json).
4. If no active sprint is found, check the current working directory for a claims.json file.
5. If still not found, stop and tell the user: "No active sprint detected. Run /init to start one."
6. Store the resolved directory as <dir> for subsequent steps.

1. Run type/evidence analysis:
2. Execute: npx @grainulation/harvest analyze <dir>
3. This returns claim type distribution (constraint, factual, estimate, risk, recommendation, feedback) and evidence tier distribution (stated, web, documented, tested, production).
4. Capture the output for the summary.

1. Run stale claim detection:
2. Execute: npx @grainulation/harvest decay <dir> --days 7
3. This flags claims that haven't been updated, corroborated, or challenged in 7+ days.
4. Capture the list of stale claim IDs and their ages.

1. Display the analytics summary:

``` Harvest: <sprint-slug> ─────────────────────────────

Claim distribution: constraint: <n> | factual: <n> | estimate: <n> risk: <n> | recommendation: <n> | feedback: <n>

Evidence quality: stated: <n> | web: <n> | documented: <n> | tested: <n> | production: <n>

Weak areas:

• <list any evidence tiers with 0 claims, or types with heavy concentration>
• <flag if >60% of claims share the same type (type monoculture)>
• <flag if >50% of evidence is "stated" or "web" (weak evidence base)>

Stale claims (<n> total):

• <id>: "<summary>" — <age> days stale
• ...

Velocity: <output from analyze, e.g. claims/day, time between phases> ```

1. Suggest full retrospective:
2. Tell the user they can generate a full HTML retrospective report:

`` For a full retrospective report: npx @grainulation/harvest report <dir> -o output/harvest.html ``

• If the user passed --full, run the report command directly and write the output to output/harvest.html.

1. Recommend calibration if sprint is complete:
2. If the sprint appears to be in a late phase (has recommendations, has a brief, or the user mentioned shipping), suggest:

`` Sprint looks complete. Score your predictions: /calibrate --outcome "what actually happened" npx @grainulation/harvest calibrate <dir> ``

• If the user passed --calibrate, run npx @grainulation/harvest calibrate <dir> and display the prediction accuracy results.

1. Suggest next steps based on the findings:

`` Next steps: /challenge <stale-id> -- revisit the stalest claim /witness <weak-id> <url> -- strengthen weak evidence /blind-spot -- check for structural gaps /brief -- compile findings into a decision doc ``

Tailor the suggestions:

• Stale claims exist -> suggest /challenge or /research on the stalest
• Weak evidence base -> suggest /witness or /research
• Type monoculture -> suggest /challenge to diversify
• Sprint looks healthy -> suggest /brief or /present