Skip to content
Back to skills

Evaluating Sprint Complete

ASecurity

Evaluate the nine-invariant sprint-complete predicate.

  • 3 stars
  • 0 votes
  • 0 copies
  • 1 view
  • Added November 16, 2026
toolsgit

Security analysis

A100/100

Pro scans all 5 files and shows the line behind each finding

Scanned November 16, 2026

npx -y skills add Acendas/shipyard --skill evaluating-sprint-complete --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Evaluating Sprint Complete?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Evaluating Sprint Complete
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/acendas-evaluating-sprint-complete/badge)](https://www.skillsdirectory.com/skills/acendas-evaluating-sprint-complete)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
name: evaluating-sprint-complete
description: Evaluate the nine-invariant sprint-complete predicate.
disable-model-invocation: true
---

# Evaluating Sprint Complete

A sprint is complete only when nine invariants hold simultaneously. This skill is the gate `/ship-execute` runs after the last wave-completion verifier passes and before declaring the sprint shippable. It composes the wave-level guarantees into the sprint-level guarantee.

## When to Invoke

`/ship-execute` Step 5 (Sprint Completion) calls this skill in the pre-review mode after:

- The full-suite test command exits 0.
- The sprint-boundary verify-probe has completed and its capture is on disk.
- The execute pipeline has completed all wave gates and is ready to hand the sprint to review.

Other entry points:

- `/ship-review` may invoke this skill with `review_verdict_path` after its scanner/review stages as a post-review release pre-flight.
- Manual invocation when a user wants to spot-check sprint completeness before declaring shippable.

## Inputs

- `sprint_id` — e.g., `sprint-003`
- `data_dir` — literal `<SHIPYARD_DATA>` path
- `working_branch` — git branch
- `sprint_base_sha` — working-branch HEAD when the sprint started (from SPRINT.md frontmatter `base_sha` or computed via `git merge-base`)
- `sprint_head_sha` — current HEAD
- `sprint_verify_capture` — path to the sprint-boundary verify-probe capture
- `sprint_verify_exit_code` — exit code from sprint-boundary verify
- `demo_probe_event_window_start` — ISO timestamp; only `acceptance_probe_completed` / `user_flow_probe_confirmed` events at or after this count for Invariant 8 (typically SPRINT.md `started_at`)
- `review_verdict_path` — path to the latest `/ship-review` verdict file (or null if review hasn't run)

## The Nine Invariants

Per-invariant checks and primitives are in [references/invariants.md](references/invariants.md); read it before running the skill.

Invariant 7 is **skipped**, not failed, on the first `/ship-execute` Step 5 invocation because `/ship-review` runs after execute. The pre-review call surfaces invariants 1–6, 8 and 9 before burning review time on a structurally incomplete sprint. Invariant 8 catches broken cross-task wiring at execute-time instead of waiting for `/ship-review` Stage 4.8.

## The Predicate Aggregation

Evaluate all nine invariants in order (the caller can pass `short_circuit: true` to stop at the first FAIL, but the default is full evaluation so the caller sees the complete picture):

```text
Run invariants 1..9.
If `review_verdict_path` is null, this is the pre-review execute invocation:
  - invariant 7 is SKIPPED and must not count as a failure;
  - invariant 3 accepts linked features in `status: in-progress`;
  - all other invariants must PASS.
Aggregate:
  All PASS         → emit sprint_complete_passed; return STATUS: COMPLETE.
  Any FAIL         → emit sprint_complete_failed with the failing invariant list;
                     return STATUS: INCOMPLETE with per-invariant detail.
```

This skill does NOT do recovery. Unlike `verifying-wave-completion` (which tries to self-heal at the wave boundary), sprint-complete failure halts cleanly — recovery is the user's call at this layer.

## Required Return Shape

The orchestrator parses these lines exactly:

```
STATUS: COMPLETE
SPRINT: <sprint_id>
INVARIANTS_PASSED: 1,2,3,4,5,6,7,8
```

OR:

```
STATUS: INCOMPLETE
SPRINT: <sprint_id>
INVARIANTS_FAILED: <comma-separated list>
DETAILS:
  invariant_1: <one-line summary if failed>
  invariant_2: <...>
  ...
```

Concrete examples of both return shapes live in [examples/](examples/).

## Event Log Emissions

The events this skill emits and their fields are catalogued in [event-types.md](../dispatching-operational-task/references/event-types.md) under "Emitted by `evaluating-sprint-complete`". Use names from the catalog rather than inventing variants inline.

## Integration With Other Skills

- **Routing.** `/ship-execute` Step 5 invokes this skill as the final gate before flipping the sprint to `status: completed`. STATUS: INCOMPLETE halts the sprint — the user sees the failing invariants and decides next action.
- **`verifying-wave-completion`** is the wave-level analog. The two compose: wave-completion guards advance-past-a-wave; sprint-complete guards declare-sprint-done.
- **`verifying-completion`** is the per-claim Iron Law (don't claim done without running the verifier). This skill IS the sprint-level verifier.
- **`/ship-review`** runs after the pre-review `/ship-execute` invocation. If this skill is invoked again post-review with `review_verdict_path`, the review verdict becomes invariant 7's input; otherwise invariant 7 is skipped by contract.

Files in this skill

  • SKILL.md4.6 KB
  • examples/return-complete.txt71 B
  • examples/return-incomplete-multi.txt864 B
  • examples/return-incomplete-pre-review.txt388 B
  • references/invariants.md11.2 KB

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

View all in tools →

Comments

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.