Skip to content

feat: add role and slug to triage harness - #4

Merged
ggallen merged 1 commit into
mainfrom
add-role-slug
Jun 30, 2026
Merged

ggallen merged 1 commit into
mainfrom
add-role-slug

Conversation

@ggallen

@ggallen ggallen commented Jun 30, 2026

Copy link
Copy Markdown
Member

Summary

  • Adds role: triage and slug: fullsend-ai-triage to harness/triage.yaml
  • Required for ADR-0058 config-driven agent registration, which resolves harnesses directly from URLs without generating thin wrappers on disk
  • Harness validation requires a non-empty role field; without it, fullsend run rejects the harness when loaded via config

Test plan

  • Verify YAML parses correctly
  • Values match scaffold harness in fullsend-ai/fullsend

🤖 Generated with Claude Code

ADR-0058 agent registration resolves harnesses directly from config
URLs without generating thin wrappers. Harness validation requires
a non-empty role field. Add role and slug so the triage harness
can be loaded via config-driven resolution.

Signed-off-by: Claude Opus 4.6 <noreply@anthropic.com>
Signed-off-by: Greg Allen <gallen@redhat.com>
@qodo-code-review

Copy link
Copy Markdown

PR Summary by Qodo

Add role and slug metadata to triage harness YAML

✨ Enhancement ⚙️ Configuration changes 🕐 Less than 5 minutes

Grey Divider

AI Description

• Add required role to enable config-driven harness validation.
• Add stable slug for URL-based harness resolution per ADR-0058.
• Ensure fullsend run accepts the triage harness when loaded via config.
Diagram

graph TD
  A["fullsend CLI"] --> B["Config URL resolver"] --> C["harness/triage.yaml"] --> D["Harness validator"] --> E["Agent registry"]
  C --> F["role: triage"]
  C --> G["slug: fullsend-ai-triage"]

  subgraph Legend
    direction LR
    _cli[CLI] ~~~ _cfg[Config] ~~~ _file[File] ~~~ _val[Validation] ~~~ _meta[Metadata]
  end
Loading
High-Level Assessment

Adding role and slug to the harness YAML is the simplest and most compatible fix for ADR-0058 config-driven resolution: it satisfies existing validation without introducing wrapper-generation or special-case logic.

Files changed (1) +3 / -0

Other (1) +3 / -0
triage.yamlAdd required role and stable slug metadata +3/-0

Add required role and stable slug metadata

• Adds 'role: triage' to satisfy harness validation when loaded via config-driven resolution. Adds 'slug: fullsend-ai-triage' to provide a stable identifier for URL-based harness discovery.

harness/triage.yaml

@qodo-code-review

Copy link
Copy Markdown

Code Review by Qodo

🐞 Bugs (0) 📘 Rule violations (0) 📎 Requirement gaps (0)

Grey Divider

Great, no issues found!

Qodo reviewed your code and found no material issues that require review

Grey Divider

Qodo Logo

@ggallen
ggallen merged commit c870dc3 into main Jun 30, 2026
3 checks passed
andrewevansrh added a commit to andrewevansrh/agents that referenced this pull request Oct 21, 2026
…ntegration

- Sanitize JIRA_TOKEN before ::add-mask:: to prevent workflow-command
  injection via crafted tokens (finding fullsend-ai#1)
- Remove reporter from required-field exemptions so validation catches
  projects that require Reporter without a default (finding fullsend-ai#4)
- Track whether Labels is required and reject proposals with empty
  labels before create instead of letting Jira reject them (finding fullsend-ai#5)
- Defer duplicate-check failure exit until after summary and Slack
  notification so already-created issues are reported (finding fullsend-ai#6)
- Distinguish absent isLast from explicit false in createmeta pagination
  so a paginated response without total is not treated as single-page
  (finding fullsend-ai#7)
- Document re-registering the agent source to the local Jira harness
  so fullsend resolves it instead of the remote pinned source (finding fullsend-ai#9)
- Pass createmeta pages through temp files instead of --argjson to avoid
  argument-size limits on large projects; strip allowedValues (finding fullsend-ai#10)
- Add _scribe_jira_api_with_status for error-body-preserving API calls
  and use it for label writes (finding fullsend-ai#11)
- Regenerate bundled scripts

Signed-off-by: anevans <anevans@redhat.com>
andrewevansrh added a commit to andrewevansrh/agents that referenced this pull request Oct 26, 2026
…ntegration

- Sanitize JIRA_TOKEN before ::add-mask:: to prevent workflow-command
  injection via crafted tokens (finding fullsend-ai#1)
- Remove reporter from required-field exemptions so validation catches
  projects that require Reporter without a default (finding fullsend-ai#4)
- Track whether Labels is required and reject proposals with empty
  labels before create instead of letting Jira reject them (finding fullsend-ai#5)
- Defer duplicate-check failure exit until after summary and Slack
  notification so already-created issues are reported (finding fullsend-ai#6)
- Distinguish absent isLast from explicit false in createmeta pagination
  so a paginated response without total is not treated as single-page
  (finding fullsend-ai#7)
- Document re-registering the agent source to the local Jira harness
  so fullsend resolves it instead of the remote pinned source (finding fullsend-ai#9)
- Pass createmeta pages through temp files instead of --argjson to avoid
  argument-size limits on large projects; strip allowedValues (finding fullsend-ai#10)
- Add _scribe_jira_api_with_status for error-body-preserving API calls
  and use it for label writes (finding fullsend-ai#11)
- Regenerate bundled scripts

Signed-off-by: anevans <anevans@redhat.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant