Skip to content

Feature request: export full context pack #152

Description

@aymenfurter

A user submitted PR #113 proposing an "Export Full Context Pack" command for AI Engineer Coach. The idea is to save a deeper Markdown/JSON context bundle (anti-pattern data, flow/context/config/workflow/session data) for use with ChatGPT or VS Code coding agents, with a Safe mode that omits raw prompt/response turns and a Full Raw mode with bounded raw session excerpts.

Logging this as product feedback: there is interest in an export capability for the full context.

Note: we are not accepting external contributions for this feature. This issue tracks the request only and is not an invitation for a PR.

Reference: #113

Activity

  1. github-actions commented on Jun 20, 2026

    @github-actions

    🎯 Triage report

    This is a sufficiently detailed feature request asking for a richer context export that could be consumed by coding agents, with explicit mention of Safe vs Full Raw export modes. The existing enhancement and wontfix labels already match both the request type and the current maintainer stance in the issue body, and I did not find a separate existing issue describing the same request.

    📊 Assessment

    Dimension Value Reasoning
    Type Feature The issue requests a new export capability rather than reporting a defect or maintenance task.
    Labels enhancement, wontfix These existing labels accurately capture the request and the stated decision not to pursue it at this time.
    Severity Low This is product feedback for a non-blocking enhancement; no regression, outage, or broken workflow is described.
    Coding agent Not suitable Implementing this would require product/privacy decisions around raw session export, redaction boundaries, defaults, and supported output shape before coding starts.
    💡 Notes and suggestions

    If this is reconsidered later, the main design questions appear to be:

    1. Export scope: exactly which entities belong in a "full context pack" (sessions, anti-patterns, context health, workflow/config metadata, raw excerpts).
    2. Privacy/safety defaults: whether Safe mode should be the default, how raw turns are redacted or bounded, and how sensitive content is handled.
    3. Output contract: whether this should extend existing summary export surfaces or ship as a separate export format optimized for downstream agents.
    4. UX/policy: whether the feature is intended for local-only workflows, and how clearly the UI should communicate the implications of Full Raw export.

    The referenced PR #113 is useful as prior design input, but I did not find a separate open issue that makes this one a clear duplicate.

    Generated by Agentic Triage for issue #152 · 29.6 AIC · ⌖ 9.63 AIC · ⊞ 22.4K · ◷

    Add this agentic workflows to your repo

    To install this agentic workflow, run

    gh aw add githubnext/agentics/workflows/issue-triage.md@df35cf29fb856d1c3b8f023ed46d19126e7813bf
    
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Labels

    enhancementNew feature or requestwontfixThis will not be worked on

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions