Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.
55e4b27385
track-gate: the txt IS the pipeline - segment fragments are now plain text (segment_NN.txt = the scene prompt verbatim + one '# handoff:' line; the approved outline row supplies id, frames, modes and images, so nothing is retyped and the old verbatim-copy failures cannot happen by construction) and the merge writes ONE file by default: prompt_override.txt for zero-media stories, prompt_override.json for stories with media (tokens are JSON-only); plan.json + track_data.json only behind --plan (the 5B patch path), prompt_override.json behind --json for zero-media continuity modes/locks. verify_track_rules gains fragment_from_text/load_text_fragment/load_fragment_any (reusing validate_fragment + prompt_rules + _check_override_refs stray-token logic on the synthesized fragment; rejects |, hand-written [start-end] markers, @, and zero-media <Picture N>) and the segment command takes .txt or .json in every position incl. --prev; 11 new suite cases, 57/57. SKILL.md/README flipped: outline.json is the only JSON the default flow ever writes; the full JSON fragment is demoted to the variant-b / custom-system-prompt escape hatch; examples/ swaps the three JSON segments + track_data.json for the Sam story's segment_01-03.txt. Proven end to end: outline_sam_park.json + the three txt fragments merge to ONE file byte-identical to the proven prompt_override_r2v_refmod.txt paste
3587e676fa
track-gate: plain text IS the deliverable - the skill told the agent to write JSON (goal line said 'PRIMARY deliverable is prompt_override.json'; smoke test, file lists, Step 4 and Step 5A all JSON-centric), so it kept writing the old format; now the paste is plain text end to end: verify_track_rules gains validate_prompt_override_text (mirror of the pack's parse_override_segments: split on |, strip, drop empties, trailing [s-e(,type)] marker lifted, @imageN/@audioN/@videoN refs vs inventory, <Picture N> and stray @ rejected, marker frames/types pinned to the approved rows, markerless form requires equal row lengths, markers all-or-nothing) + build_plain_text_override (fragment text VERBATIM - newlines survive, six-section structure intact - not the JSON's collapsed prompts; bare-| join, byte-identical to the user's proven node186 paste) and the override command takes .txt or .json; build_track_plan writes prompt_override.txt for EVERY zero-media story (markers auto-added for uneven lengths / non-t2v types), media stories keep JSON-only with an explicit note; SKILL.md/README/schema/wiring docs flipped: plain text default, JSON demoted to the <Picture N>/continuity/locks option; suite 46/46; merge output proven byte-identical to prompt_override_r2v_refmod.txt
2cf162c451
SKILL.md Step 5: the chain plan targets the standalone Context Loop pack only - ComfyUI-Easy-Media has no plan_json/shots consumer (diagnosed live: pasted plan ignored, default single task segment 0-240 frames renders the whole story as one 10s clip); point Easy-Media users at easy-media-track-gate
0723a2dbd4
Fix refmod isolation false positive: 'like' joins REF2VA_STOPWORDS - any refmod phrased 'dresses like ...' banned the common function word from every scene body (live test run: 'cries like an infant' rejected with found: like). No behavior change for distinctive tokens; embedded suite unchanged at 7,712 B, example plan PLAN OK
bbd46a4764
track-gate: worked r2v+refmod plain-text example - examples/prompt_override_r2v_refmod.txt is the Sam park story as one prompt_override paste: three six-section r2v scene prompts, Sam refmod verbatim in every subject_definitions, pipes between, [0-240,r2v]/[240-480,r2v]/[480-720,r2v] markers at segment ends; proven through parse_override_segments + build_multitrack_data_from_prompt_override: 3 tasks TYPE r2v, frames [0-240]/[240-480]/[480-720], task-duration sum 720 (that sum, not the builder's 721 metadata, drives the MiniMax total), text byte-identical to plan shots, marker stripped clean; schema doc: marker end-placement, newline survival, pointer to the example
e3e66dd628
track-gate: plain text is the primary prompt_override form (user-verified) - the input splits 'scene1 | scene2 | scene3' on pipes with no JSON: parse_override_segments, even slices of the editor's total_length, trailing | a no-op, TYPE t2v; [s-e,type] and @mediaN are text-path extras, <Picture N> is JSON-only; schema doc now leads with the text form and demotes the JSON to per-segment control; wiring doc 'two forms' section; merge writes prompt_override.txt for zero-media equal-duration stories; suite 34/34
1473abcbd2
track-gate: prompt_override minimized to the parser's real contract - parse_minimax_prompt_override requires ONLY type+prompts; version is read by nothing (validator no longer demands it); duration/generation_type/continuity_mode/locks emitted only when non-default (code defaults: '10' SECONDS, r2v, shot, 0); proven with the pack's own code: minimal 3-key Sam payload builds byte-identical TRACKS_INFO vs the old 9-key one; schema doc rewritten around the minimal form, no frame fields
20143da60e
track-gate: relay single-pass OOMs at 720 total frames (user-verified) - demote task_index=-1+promptRelay to <=~360-frame timelines; per-clip task_index=0/1/2 with promptRelay kept is the production path (relay-at-index formats the single clip's [s-e] range, verified in basic.py); TYPE line updated for the r2v flip
9b197ae024
track-gate: RefMod workflows author segments as task_mode ref (generation_type r2v) even with no timeline images - user-verified: refmods pair with reference-mode conditioning; Sam payload flipped default->ref, regenerated, validator green
b45e6b72e7
track-gate: single-pass timeline render proven - prompt_override needs the type field, task_index=-1 + prompt_format=promptRelay renders all scenes in one pass ([s-e] range relay prompt, 17k+5 snap 720->719); corrects the earlier -1-is-not-a-render-path claim; proven against my timeline ref.json with the repo's own code
f6be7ffcc5
Refmod verbatim is now machine-checked: outline_refmods() extracts the story's identity/voice block from the outline's subjects strings and diffs every scene's subject_definitions line against it (fragment --outline + the merge); rewrites, form swaps and invented refmods fail with actionable errors; embedded suite gains the gate fixtures and stays under Cline's 8k output cap (7,712 B)
dc8d9fab81
Skill is install-location independent: SKILL_DIR concept (loaded-from folder) replaces project-root assumption; all command forms use %USERPROFILE%\.agent\skills\h3-plan-gate; generated files go to the working folder; README install section rewritten for global skills dirs (.clinerules/.cline kept as repo-workspace extras)
36c2dbf46e
Cline-aware bounded reads: cite exact line ranges (PROMPT A 53-156, PROMPT B 157-568, single-pass fence 30-428), run-don't-read the 62KB validator, chunking fallback for truncated reads - Cline caps reads at ~50k chars and cuts sooner on small model contexts
c0d377b894
build_plan.py delivery print and docstring now route the paste to the internal plan_json widget (dynamicPrompts:False) - stops the LLM from steering the user back into a multiline/dynamic-prompt input
23d0def305
Field-verified: multiline text input strips braces, same input's single-line mode passes probe + full plan - the probe decides, not the label
dbb6c82e1f
Ship the paste-surface probe and the one-line example plan as repo files (plan_oneline.json, brace_probe_oneline.json) - SKILL.md references the probe file
99c3e89beb
Step 5 corrected: internal plan_json widget is the only safe paste surface - ComfyUI frontend dynamic-prompt serializer strips braces from dynamicPrompts:true widgets at queue time (Extra data fingerprint = quoted first key length; probe calibration included)