atomicinnovation Show Jira Issue
!accelerator config context --skill show-jira-issue --fail-safe
Fetch and render a single Jira issue by key. Work through the steps below in order.
Step 1: Parse the issue key and flags
The first positional argument is the issue key (e.g. ENG-42, PROJ-1234).
Remaining arguments are passed through to the subcommand as-is.
--render-adf defaults to ON — single-issue reads are for humans, so rendered
Markdown is the natural output. Pass --no-render-adf verbatim if the user
explicitly asked for raw ADF or JSON.
Step 2: Fetch the issue
Run the show subcommand with the key and any flags supplied:
accelerator jira show <ISSUE-KEY> [flags]
Run the bare launcher directly as an executable; never prefix it with
bash/sh/env and never pipe its output (a wrapper prefix or a pipe escapes
the skill's allowed-tools permission and forces an unnecessary prompt).
The subcommand emits a JSON document with a top-level outcome keyword:
found or not-found. On not-found, tell the user the issue was not found
and check the key. On a non-zero exit, show the error — a credential or site
failure names an E_* cause on stderr; suggest /init-jira.
Step 3: Render the result
Parse the JSON response and present a human-readable summary from .fields:
- Heading:
## KEY — Summary text - Fields block: Status, Type, Priority, Assignee, Reporter (omit if absent).
- Description: render inline as Markdown (already rendered when
--render-adfis on). - Comments (when
--comments Nwas passed): render each comment as a mini conversation block — author, timestamp, body.
If the fields object is sparse (e.g. --fields summary,status was used),
only render the fields that are present; do not invent missing ones.
Examples
Example 1 — look up an issue User: "look up PROJ-1234" Skill invokes:
accelerator jira show PROJ-1234
Then renders the issue with description as Markdown.
Example 2 — show recent comments User: "what's the discussion on ENG-42 — show me the last few comments" Skill invokes:
accelerator jira show ENG-42 --comments 5
Then renders the summary + last 5 comments as an inline conversation.
Example 3 — raw JSON escape hatch User: "give me the raw JSON for ENG-42 — I'm piping it to jq" Skill invokes:
accelerator jira show ENG-42 --no-render-adf
Then prints the response with ADF intact.
!accelerator config instructions show-jira-issue --fail-safe