Wayland Morning Report

Morning Report

Estimated time: about two minutes for a 30-name watchlist.

This workflow reads the user's live TradingView chart through the TVControl connector and writes a brief. It computes nothing. Every number in the output is a number their own chart printed, so they can point at any line and find it on screen.

Ground rules

• Write everything to the deliverables directory named in your run instructions, and never outside the workspace — everything outside the workspace is refused by the sandbox. Your run instructions name that directory as an ABSOLUTE path, and it is the destination. Writing anywhere else means the run publishes nothing at all.
• That directory is a destination, not an address. On a scheduled run it is a staging directory and it is deleted the moment the run publishes — the app publishes by renaming it, so the path you wrote to stops existing at that instant. Write to it and never quote it back to the user. Naming a file is fine; naming its directory is telling the user to go somewhere that is gone.
• Do not read WAYLAND_OUTPUT_DIR. It is set on the engine process and the engine does not forward it to shell commands, so it always resolves EMPTY and every ${WAYLAND_OUTPUT_DIR:-…} fallback silently wins. If an older instruction anywhere tells you to read it, ignore that instruction.
• Pin the directory to a shell variable before you write anything. Create it if it does not exist. If it contains a .git folder, or sits inside one, stop and ask for a different path rather than dirtying somebody's repo on a schedule.
• Never fabricate numbers. Every figure must come off the chart you read. If the chart could not be read, the brief says so and carries no figures at all.
• Never claim success you did not verify. A scheduled run is marked delivered the moment this prompt arrives, long before anything has run. The only place the user learns whether the report actually worked is what you write in this thread, so state the outcome explicitly every time.

Steps

Step 1: Prove the chart is answering (uses: morning-prep)

Do NOT assume you hold the TVControl connector - find out. Search your tool catalogue for tv_health_check and call it. Only an empty search is evidence of absence.

Whether the tools are there depends on how this run started, and the difference matters when you report the outcome:

• Run on demand, in a chat where the user has TVControl connected, you hold the chart tools.
• Run on a schedule, you currently hold NO MCP connectors at all. Every user connector is deliberately scoped out of an unattended run, so an empty catalogue on a 07:00 run is the expected state, not a fault.

Three outcomes, and each has one honest answer:

• The tools are not there. Say which case you are in: on a scheduled run, that scheduled runs do not currently receive the chart connector, so no brief can be produced and the user should run it on demand instead; on an on-demand run, that TVControl is not installed or not connected. Either way write that and stop. Produce no market content, and do not describe this as a chart or TradingView failure - it is neither.

• datafeed.state is disconnected. Every number on that chart may be stale, the indicators included, because they are computed off the same bars. Write that, and read no figures.

• Healthy. Continue.

• Output: a health verdict you actually received

• Key focus: fail with a clear sentence, never with a stack trace

Step 2: Resolve the watchlist by name, never by "active"

Call watchlist_list. It returns every list with its id and symbol count.

Do not use watchlist_get: it reads whatever TradingView currently marks active, which has been measured naming a 29-symbol list while the chart in front of the user held a 74-symbol one, and reporting that as a success.

Pick the list the user configured, by id. If the run has no configured list and more than one exists, write a brief saying which lists were found and that the user needs to choose one. Do not guess.

Step 3: Read the chart across the watchlist

Use batch_run with the get_pine_tables action and the whole symbol list in ONE call. It restores the starting chart state when it finishes.

For each result, confirm the panel's own stamp names the symbol you asked for. A panel that still names the previous symbol has not repainted, and its numbers belong to a different instrument — that is a FAILED read for that symbol.

read + failed must equal the number of symbols you asked for, every time. A symbol is never dropped, never back-filled, and never quietly counted as neutral.

Step 4: Write the brief and report the outcome

Pin the directory, then write the file. Substitute <deliverables_dir> with the absolute deliverables directory your run instructions name:

OUT="<deliverables_dir>"; mkdir -p "$OUT"

Write a standalone morning-brief.html into "$OUT". It must state, at the top: the watchlist name and its size, the timeframe, the indicators found on the chart, and the count read versus attempted.

Then report in this thread:

• how many symbols were read, out of how many attempted
• which watchlist, by name and size
• every symbol that could not be read, by name
• the outcome in one word: complete, partial, or empty

Name the file morning-brief.html when you refer to it. Do not print the deliverables directory — it is a staging directory on a scheduled run and it stops existing the moment the run publishes, so a user who follows it arrives nowhere.

An empty run is a valid outcome and must be reported as one. A brief with no figures, clearly labelled, is worth more than a brief that looks complete and is not.