Re-evaluate build-artifact gitignore protection after harness slimming
Maintainers usually reply within 1 day
Nobody has claimed this yet.
Assessment
- Difficulty
- 5/5
- Estimated time
- Over a week
- Newbie friendliness
- 35/100
- Issue type
- Feature
- Clarity
- Needs clarification
- Activity status
- Active
- Tech stack
- go
- Domain
- build-system, tooling
Research direction
Start in harness/cmd/migration-harness/main.go, especially the removed EnsureGitignore path and fetchAndWriteAnalysis. Review PR #186 and the base image or always-loaded rule options to decide which protection model is intended. Done means resolving the policy and ensuring .konveyor/ has ignore coverage, with the broader artifact behavior documented or implemented consistently.
Written by the indexing model from the issue text.
Description
Context
In PR #186 (slimming the harness to a minimal entry point), EnsureGitignore was removed from harness/cmd/migration-harness/main.go, and the harness no longer injects .gitignore entries for build artifacts before the agent runs.
Dylan Murray (@dymurray) flagged this in review:
Removing
EnsureGitignorehere drops the only guard against generated artifacts (.goose/,target/,node_modules/, build output) landing in commits. This matters more under the generic entry-point model, not less: a third-party operation/skill that runsgit add -Ainstead of targeted adds will push build junk onto the branch. [...] consider keeping a minimal ignore write, or moving these patterns into a committed.gitignorein the base image / an always-loaded rule, rather than relying solely on skill prompt discipline. (Regression class previously fixed in changelog 92-skill-commit-hygiene / 110-skip-unnecessary-grounding-commit.)
— https://github.com/konveyor/agentic-controller/pull/186#discussion_r3874048353
He later added a follow-up on the same thread expressing uncertainty about his own concern:
Reviewing this comment I don't know if I agree with this; I'm skeptical this comment is correct
— https://github.com/konveyor/agentic-controller/pull/186#discussion_r3874085550
That follow-up went unanswered, so this is an open question rather than a settled ask.
Related: with the grounding commit and gitignore write both gone, .konveyor/analysis.json (written by fetchAndWriteAnalysis in harness/cmd/migration-harness/main.go) is now untracked with no ignore coverage. A broad git add -A from a skill would sweep it (and .konveyor/handoff.md) into a PR.
With both the grounding commit and the
.gitignorewrite gone,.konveyor/analysis.jsonis now an untracked file with no ignore coverage. Fine within a single run (the agent reads it in-tree), but a broad add sweeps it — and.konveyor/handoff.md— into the PR. If grounding data is meant to stay entry-point-provided and never committed, add.konveyor/to a committed ignore so accidental inclusion is impossible. Same root cause as theEnsureGitignoreremoval above.
— https://github.com/konveyor/agentic-controller/pull/186#discussion_r3874048362
Ask
Decide whether the entry point should:
- Rely entirely on skill-authored
.gitignore/targeted-add discipline (current state), or - Restore a minimal, committed ignore baseline (e.g.
.konveyor/,.goose/,target/,node_modules/) in the base image or an always-loaded rule, so accidental inclusion viagit add -Aisn't possible.
At minimum, .konveyor/ should have ignore coverage regardless of the broader decision, since it's entry-point-provided data that's never meant to be committed.
- Dominant language
- Go
- Stars
- 5
- Forks
- 13
- Avg merge
- 3d 6h
- Merged PRs (30d)
- 47
Getting set up
- Ships a Dockerfile or Docker Compose file
- Has a pull request template
- No contributing guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
More from konveyor/agentic-controller
-
bug stream/controller
Difficulty 2/5 1-3 hours Newbie friendliness 76/100
konveyor/agentic-controller#164 ·
Maintainers usually reply within 1 day
-
documentation good first issue
Difficulty 2/5 1-3 hours Newbie friendliness 75/100
konveyor/agentic-controller#124 · 1 comment ·
Maintainers usually reply within 1 day
-
Difficulty 5/5 Over a week Newbie friendliness 25/100
konveyor/agentic-controller#341 ·
Maintainers usually reply within 1 day
-
Difficulty 4/5 3-5 days Newbie friendliness 48/100
konveyor/agentic-controller#323 ·
Maintainers usually reply within 1 day
-
Provider-aware Gateway verification for aws-bedrock / gcp-vertex-ai, so bad credentials fail at verify rather than mid-runPossibly taken @ibolton336 claimed this 5 days ago. Openenhancement stream/controller
konveyor/agentic-controller#304 · 1 assignee ·
Maintainers usually reply within 1 day
All issues in konveyor/agentic-controller
Similar issues
-
bug
Difficulty 2/5 1-3 hours Newbie friendliness 64/100
Maintainers usually reply within 1 day
-
bug raised-by:verifier
Difficulty 2/5 1-3 hours Newbie friendliness 76/100
medici-finance/assay#5697 · 1 comment ·
Maintainers usually reply within 1 day
-
Difficulty 2/5 1-3 hours Newbie friendliness 82/100
sonatype-nexus-community/terraform-provider-sonatypeiq#101 ·
Maintainers usually reply within 1 day
-
area: metrics area: tracing contribfest enhancement good first issue
Difficulty 2/5 1-3 hours Newbie friendliness 83/100
open-telemetry/opentelemetry-ebpf-instrumentation#4188 ·
Maintainers usually reply within 1 day
-
Difficulty 2/5 1-3 hours Newbie friendliness 74/100
blinklabs-io/dingo#6074 ·
Maintainers usually reply within 1 day