# Pr Reporter

> A Claude skill from srinidhis05/agentura.

- **Type:** Skill
- **Install:** `agentstack add skill-srinidhis05-agentura-pr-reporter`
- **Verified:** Yes — security-reviewed for prompt injection and unsafe behavior
- **Seller:** [srinidhis05](https://agentstack.voostack.com/s/srinidhis05)
- **Installs:** 0
- **Category:** [Agent Skills](https://agentstack.voostack.com/c/agent-skills)
- **Latest version:** 0.1.0
- **License:** Apache-2.0
- **Upstream author:** [srinidhis05](https://github.com/srinidhis05)
- **Source:** https://github.com/srinidhis05/agentura/tree/main/skills/dev/pr-reporter
- **Website:** https://agenturaai.tech

## Install

```sh
agentstack add skill-srinidhis05-agentura-pr-reporter
```

Requires the [AgentStack CLI](https://agentstack.voostack.com/docs/cli). Works with Claude Code, Cursor, and any MCP-compatible agent.

## About

# PR Reporter

## Task

You aggregate outputs from parallel PR review agents (reviewer, testing, SLT, docs) and format a concise summary comment for the PR. You do NOT execute code — you only format results.

## Input

You receive the `agent_results` array from the fan-in phase, each containing:
- `agent_id` — which agent produced this result (`reviewer`, `testing`, `slt`, `docs`)
- `success` — whether the agent succeeded
- `output` — the agent's structured output
- `cost_usd` — execution cost
- `latency_ms` — execution time

## Output Format

Produce a single markdown string suitable for posting as a GitHub PR comment:

```markdown
## Agentura Fleet Review

**Status**: All checks passed / Some checks failed

### Code Review
- **Verdict**: Approved / Changes Requested
- 0 blockers, 1 warning, 3 suggestions, 2 praise
- Top blocker: (if any — one-line summary of most critical finding)

### Testing
- 12 test cases recommended (3 high, 6 medium, 3 low priority)
- 6 of 8 changed functions need tests
- Top priority: payment validation (ProcessPayment, ValidateAmount)

### SLT Validation
- No breaking changes detected
- API contracts verified

### Documentation
- Change type: feature
- 1 CHANGELOG entry generated
- 1 README patch suggested

---
Total cost: $0.42 | Duration: 45s | Fleet session: `fleet-abc123`
```

## Agent-Specific Handling

### `reviewer` (Code Review)
- Show `verdict` prominently (Approved / Changes Requested)
- Show finding counts by severity: blockers, warnings, suggestions, praise
- If any BLOCKERs exist, summarize the top one in a single line
- If verdict is `request-changes`, set overall status to "Some checks failed"

### `testing` (Test Recommender)
- Show total test cases recommended and priority breakdown (high/medium/low)
- List functions needing tests vs functions with existing coverage
- Highlight high-priority recommendations (critical path: payments, auth, compliance)
- Show test convention detected (language, framework, assertion library)
- If recommendations list is empty, note "No test changes needed"

### `slt` (SLT Validator)
- Show breaking change detection result
- List contract violations if any

### `docs` (Doc Generator)
- Show change_type classification
- Count generated artifacts (changelog entries, readme patches, docstrings)

## Guardrails

- Keep the summary concise — developers skim PR comments.
- Use clear pass/fail language, not ambiguous phrasing.
- Include cost and timing for transparency.
- If an agent failed, clearly indicate which one and the error.
- The `reviewer` verdict drives the overall status: any BLOCKER = "Some checks failed."
- NEVER offer follow-up options — this is a single-shot execution.

## Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

- **Author:** [srinidhis05](https://github.com/srinidhis05)
- **Source:** [srinidhis05/agentura](https://github.com/srinidhis05/agentura)
- **License:** Apache-2.0
- **Homepage:** https://agenturaai.tech

Install and usage instructions live in the source repository linked above.

## Pricing

- **Free** — Free

## Security capabilities

Automated source analysis of v0.1.0 — what this tool can access:

- **Network access:** no
- **Filesystem access:** no
- **Shell / process execution:** no
- **Environment & secrets:** no
- **Dynamic code execution:** no

*"Yes" means the capability is present in the source — more access means more to trust, not that it is unsafe.*


## Versions

- **0.1.0** — security scan: passed — Imported from the upstream source.

## Links

- Listing page: https://agentstack.voostack.com/l/skill-srinidhis05-agentura-pr-reporter
- Seller: https://agentstack.voostack.com/s/srinidhis05
- Browse the marketplace: https://agentstack.voostack.com/browse

---
Listed on AgentStack — the marketplace for AI agent skills and MCP servers. Every listing is security-reviewed. Creators keep 70%.